Surfactants, along with antenatal steroids and assisted ventilation, have dramatically improved neonatal outcomes. The first case series, published in The Lancet in 1980, reported resolution of chest x ray findings and reduction in ventilator settings in 10 preterm babies with severe respiratory distress syndrome (RDS) who were given a modified bovine surfactant.1 In the mid 1980s the first randomised trials showed dramatic reductions in death and pulmonary air leaks when using either natural2 or synthetic surfactants.3 4 Since then there have been many good quality randomised controlled trials and meta-analyses of these trials designed to determine which surfactant is best and the optimal timing of surfactant dosing in babies with RDS. As a result, we can be confident about many recommendations for best practice, and consensus guidelines have been produced in several countries including the UK,5 USA6 and Canada,7 and in Europe.8 However, there are some areas where controversy still exists.

The aim of this review is to summarise what we know about surfactant therapy in 2009 and to highlight some of the areas where there is still lack of clarity.

SURFACTANT FOR RDS

RDS is a condition of pulmonary insufficiency that in its natural course commences at birth and increases in severity over the first 2 days of life. If left untreated death can occur from progressive hypoxia and respiratory failure. RDS is due to a lack of alveolar surfactant along with structural immaturity of the lung and it mainly affects preterm babies. The incidence increases with decreasing gestation, with about 25% of babies being affected at 34 weeks’ gestation and up to 80% of babies being affected at less than 28 weeks’ gestation. Clinically, RDS presents early with respiratory distress including cyanosis, grunting, retraction and tachypnoea. Respiratory failure may develop and the diagnosis can be confirmed on chest x ray with a classical “ground glass” appearance and air bronchograms.

The majority of clinical trials of surfactant therapy have been for the management of babies with RDS. The main function of pulmonary surfactant is to lower alveolar surface tension and prevent end expiratory alveolar collapse. Surfactant deficiency causes decreased lung compliance, a reduction in lung volume, ventilation-perfusion mismatching and raised pulmonary vascular resistance. Exogenous surfactants are therefore used to reverse this process, increasing compliance, stabilising the alveoli, reducing pulmonary vascular resistance and improving ventilation–perfusion mismatching. Many of these effects can also be achieved by using continuous positive airways pressure (CPAP). Applying a positive distending pressure, even in the face of surfactant deficiency, can prevent end expiratory alveolar collapse, reduce the work of breathing and reduce ventilation–perfusion mismatch.9 CPAP was not in such widespread use when many of the surfactant clinical trials were undertaken and antenatal steroid administration rates were also much lower than today. Advocates of CPAP have demonstrated that many babies who traditionally would have been selected for surfactant in the original clinical trials can now apparently be managed perfectly adequately using CPAP alone.10 In some institutions even babies of gestational age as low as 25 weeks can be managed on CPAP, with only a third needing surfactant and less than 10% needing ventilation within the first week of life.11 There is a strong argument for the avoidance of unnecessary intubation and mechanical ventilation solely for surfactant administration as these may potentially induce lung injury which can lead to bronchopulmonary dysplasia (BPD).12

The aim of management of RDS is to provide interventions that will maximise the number of survivors and decrease morbidity while minimising the potential adverse effects of therapy. The statements below are common to most guidelines as they are derived from meta-analyses of randomised trials and can be considered as evidence based.

Surfactant therapy for RDS improves survival and reduces pulmonary air leak

It is now abundantly clear that surfactant therapy should play a role in the management of babies with RDS. Clinical trials of artificial and natural surfactants, given both prophylactically and as rescue therapy, have consistently shown that surfactant therapy improves survival and reduces the risk of pneumothoraces.2^–4 However, these clinical trials took place in the 1980s and 1990s, an era when antenatal steroid administration was generally low. Management of RDS largely consisted of head box oxygen with mechanical ventilation for babies in whom respiratory failure developed. Use of early CPAP was mainly confined to Scandinavian countries and other forms of mechanical ventilation were not as highly refined as today. The babies in these original trials were usually less than 30 weeks’ gestation or less than 1250 g birth weight. Meta-analysis of trials of synthetic surfactant showed a reduction in BPD defined as oxygen dependency at 28 days (RR 0.75; 95% CI 0.61 to 0.92) when given as rescue4 but not when given as prophylaxis3 (RR 1.06; 95% CI 0.83 to 1.36). This is counterintuitive as it suggests that exposing babies to prophylactic synthetic surfactant, while saving lives, may increase the risk of BPD. Whether this is true of natural surfactant preparations is in doubt as the Cochrane meta-analysis shows a non-significant reduction in BPD2 (RR 0.93; 95% CI 0.80 to 1.07), but for a porcine surfactant there is a significant decrease in BPD in survivors at 28 days (OR 0.54; 95% CI 0.34 to 0.86) when prophylaxis is compared to later surfactant treatment in a meta-analysis of three trials.13

Exogenous surfactant has no apparent direct adverse effects. The main concern about surfactant instillation is related to the fact that it requires intubation and positive pressure ventilation for its distribution. Even a very short period of vigorous manual ventilation can result in significant lung injury in immature lambs.14

Surfactant preparations need to be administered by bolus directly into the lungs, necessitating intubation and a short period of ventilation for distribution

Surfactants need to be administered directly into the lungs. This means that administration involves intubation and at least a short period of assisted ventilation. All of the current recommendations emphasise the need for babies who may develop RDS to be born in centres where personnel and equipment are available to manage them properly. Intubation and mechanical ventilation are independent risk factors for the development of BPD15 and protocols for RDS management usually aim to minimise the number of babies exposed to ventilation. Use of the INSURE technique (Intubate - Surfactant - Extubate to CPAP)16 has minimised the duration of mechanical ventilation for babies with RDS being treated with CPAP and the earlier in the course of RDS this technique is applied, the greater the success in avoiding subsequent ventilation. However, the studies performed so far have not been sufficiently powered to determine if this practice reduces BPD. Animal studies have shown that surfactant distribution is best when surfactant is administered as a bolus rather than trickled in over several minutes.17 However, there is no evidence to support the practice of aiding surfactant distribution by moving babies into different positions during administration. Attempts to administer surfactant by nebulisation have not so far been successful.18

The earlier surfactant is given in the course of RDS, the better it works and some babies at very high risk of RDS should be given prophylactic treatment

Trials published during the early 1990s confirmed that surfactants were more effective at reducing death and lung injury if given earlier in the course of RDS.13 19 Several studies confirmed that surfactant prophylaxis before the clinical onset of RDS is better than waiting until RDS is firmly established.20 Prophylactic surfactant reduces mortality (RR 0.61; 95% CI 0.48 to 0.77) and pneumothorax (RR 0.62; 95% CI 0.42 to 0.89) but not BPD (RR 0.96; 95% CI 0.82 to 1.12),20 with the exception of survivors to 28 days treated with a porcine surfactant.13 Most trials compared giving surfactant within the first 15 min after birth with giving surfactant when the babies reached certain pre-defined criteria such as requirement for mechanical ventilation or Fio₂ above 0.40, and the timing of treatment in the rescue groups was relatively late, between 1.5 and 4–6 h. A large randomised trial addressed the issue of whether surfactant should be given before the first breath by comparing two methods of prophylaxis with a bovine surfactant.21 In this study of 651 babies, there was no difference in clinical outcome for babies randomised to immediate surfactant prophylaxis compared with surfactant prophylaxis given at about 10 min of age after stabilisation and clinical confirmation of correct placement of the endotracheal tube. Unfortunately, the downside to surfactant prophylaxis is exposure to intubation. The randomised trials were undertaken in an era of low antenatal steroid use, and even then, about 50% of babies randomised to receive later surfactant did not need it after all. Apart from cost considerations, many neonatologists are uncomfortable intubating and ventilating preterm babies for prophylactic surfactant administration in the current era of increased antenatal steroid usage and CPAP. It is not yet clear whether true prophylaxis is superior to an early rescue treatment strategy of administering surfactant as soon as the baby begins to develop clinical signs of RDS. Of the studies included in the Cochrane meta-analysis, the earliest median time of administration in the rescue surfactant group was 1.5 h after birth. In this study there were clinical advantages for the babies given prophylaxis.22

Natural surfactant preparations are superior to synthetic surfactant preparations

When surfactants were first introduced during the 1980s there were more commercial products available than there are today. Synthetic surfactants such as colfosceril palmitate (Exosurf) and pumactant (ALEC) were produced by mixing dipalmitoylphosphatidylcholine, the main phospholipid found in endogenous surfactant, with spreading agents such as hexadecanol and tyloxapol or phosphatidylglycerol. Natural surfactants were produced by extraction from animal lungs: beractant (Survanta) from cows and poractant alfa (Curosurf) from pigs being the two that are available in the UK. Natural surfactants have a much more rapid onset of clinical effects than synthetic surfactants and comparative clinical trials during the 1990s were able to quantify this. A Cochrane meta-analysis of 11 of these trials showed the superiority of natural surfactants, with reductions in death (RR 0.87; 95% CI 0.76 to 0.98) and pneumothorax (RR 0.63; 95% CI 0.53 to 0.75), although there was no significant effect on BPD.23 Synthetic surfactants have been withdrawn from the market in the UK and only two natural surfactant preparations are currently available.

There is a move to produce a synthetic surfactant which can match the natural surfactants for efficacy but can be produced more cheaply. There are theoretical concerns regarding the use of animal products, although to date they have not been shown to have any adverse immunological or infectious complications.24 25 Synthetic peptides or proteins which mimic the actions of surfactant proteins may provide the answer. To date, only lucinactant has reached the stage of being used in comparative clinical trials in preterm infants.26 27 Although this newer synthetic surfactant appears superior to the older synthetic surfactants, the comparative studies with natural surfactants were criticised for early trial closure and inadequate sample sizes,6 28 and so far this drug has not been licensed for use in preterm babies.

Although there have been some comparative studies of different natural surfactants, it is not clear if the apparent superiority of the porcine compared to the bovine product is dose related rather than due to chemical or physical differences between the surfactants (see below).

CURRENT DILEMMAS IN SURFACTANT TREATMENT FOR RDS

Although much of what we do in terms of neonatal respiratory care is evidence based, there are still some areas of controversy. These areas of uncertainty are outlined below.

Prophylaxis or very early rescue surfactant

Most of the present surfactant guidelines still recommend prophylactic surfactant for babies at the highest risk of RDS, such as those less than 27 weeks’ gestation or those less than 29 weeks’ gestation who have not had antenatal steroids because of the survival advantage shown in randomised trials.20 However, there is still some uncertainty and clinical trials are being undertaken to see if babies in the present era of increased antenatal steroid and CPAP use can be managed without resorting to intubation unless it is clear that they have or will develop significant RDS. We currently do not know if there are any real advantages to prophylaxis over very early rescue therapy at say 1 h of age, and many believe that intubation for surfactant should be deferred until it is clearly necessary. Neonatal units which have adopted this practice seem to have reduced rates of BPD with no increase in mortality.10 It is possible to determine the absence of surfactant in individual babies with the stable microbubble or click tests, but they are not widely used at present, perhaps because RDS is often clinically apparent before the results of tests become available.29

Where the practice of routine intubation of spontaneously breathing preterm infants has been critically observed, it is clear that it often does not go smoothly, with frequent multiple attempts at intubation and episodes of profound apnoea and bradycardia.30 CPAP is now used frequently in most NICUs and can be applied immediately after birth, either by a nasopharyngeal tube or nasal prong, helping babies to establish a functional residual capacity without resorting to immediate intubation or positive pressure ventilation. Babies who develop significant RDS can then be given early rescue surfactant as a semi-elective procedure. There have been several recent studies aimed at clarifying these issues of early respiratory management of very preterm infants. The COIN Trial randomised 610 babies between 25 and 28 weeks’ gestation, who were breathing spontaneously at 5 min of age, to either early CPAP alone or prophylactic intubation with planned early extubation to CPAP. Surfactant was given at the discretion of the attending neonatologist, so this was a trial of early CPAP rather than early surfactant. Overall, 46% of the CPAP group went on to require intubation within the first 5 days. More babies developed pneumothoraces in the CPAP group (9% vs 3%; p = 0.001). Although there was no significant difference in the number of deaths or the rate of BPD at 36 weeks’ gestation, the trend favoured the CPAP group.31 Recently a study from Columbia showed benefits of surfactant therapy within 1 h of birth for babies of 27–31 weeks’ gestation who had early RDS and were managed on CPAP.32 The surfactant treated babies had less need for ventilation, fewer pneumothoraces and a trend towards reduced BPD. The CURPAP study is a European multicentre trial designed to determine if administration of prophylactic surfactant to babies between 25 and 29 weeks’ gestation who are started on CPAP from birth is better than early rescue surfactant in terms of reducing the need for mechanical ventilation over the first 5 days of life.33 The results, although not yet published, suggest at best only modest benefits of this procedure.

At present it still seems reasonable to administer prophylactic surfactant to extremely preterm infants at high risk of RDS, although where this gestational age cut-off should be applied may vary between units according to local policy.8 When deciding who to treat, consideration should also be given to whether or not the mother received antenatal steroid treatment.

When should a second and third dose of surfactant be given?

Another issue which lacks clarity is the timing of repeat doses of surfactant. A Cochrane review shows that a strategy of allowing multiple doses of natural surfactant rather than a single dose further reduces the risk of pneumothorax (RR 0.51; 95% CI 0.30 to 0.88) and there is also a trend towards reduction in mortality.34 However, the studies included in this review were conducted in the early 1990s when antenatal steroids were used sparingly and they also included relatively mature infants. The first dose of surfactant was given comparatively late about 6–12 h after birth and the policies at the time would have been to use longer periods of mechanical ventilation rather than CPAP. Manufacturers of natural surfactants make specific recommendations regarding re-treatment: beractant (Survanta) may be repeated within 48 h at intervals of at least 6 h for up to four doses; poractant alfa (Curosurf) may be repeated 12 h later for two further doses if still intubated, and after prophylaxis may be repeated 6–12 h later and after a further 12 h. Recently, a large trial of 1267 babies who met re-dosing criteria (Fio₂ more than 0.30) were randomised to receive a second dose of bovine surfactant or wait until the Fio₂ was more than 0.40.35 Babies with uncomplicated RDS fared no worse when re-dosed at this higher threshold, although about one quarter were sicker babies with “complicated RDS” (perinatal compromise or sepsis) and this group had lower mortality when re-dosed at the lower threshold. The Canadian Guideline makes a fairly specific recommendation that babies should be re-treated if they remain in more than 30% oxygen as early as 2 h after the first dose.5 The European Guideline is less specific, recommending re-treatment if there is ongoing evidence of RDS such as the need for mechanical ventilation and supplemental oxygen.8 For babies on CPAP, re-treatment is recommended if they need more than 50% oxygen or if they seem to be likely to need mechanical ventilation.

Which surfactant should we use?

The other dilemma facing neonatologists is which natural surfactant to choose. In the UK we have a choice of poractant alfa (Curosurf) and beractant (Survanta). In the USA calfactant (Infasurf), another bovine surfactant, is also available, in Canada they have BLES (bLES) and in some countries in Europe bovactant (Aveofact) is also used. It is difficult to obtain information about the other natural surfactants available around the world, including Surfactant-TA in Japan, Surfacen in Cuba and Newfacten in Korea. All of these natural surfactant preparations are slightly different in terms of the concentrations of phospholipids and surfactant proteins (table 1).36 There have been at least 10 randomised trials aimed at determining if there are clinical differences between these various natural surfactant preparations.37^–39 The largest trials have compared beractant and calfactant, with one study recruiting over 2000 babies.40 However, the planned sample size was not reached due to slow recruitment, but in the babies that were studied, there was no significant difference in the number of babies surviving without BPD, in either the prophylaxis arm or the rescue arm of the study. Studies comparing poractant alfa and beractant are smaller, but they consistently show that the porcine surfactant improves oxygenation more rapidly. In a single trial of 293 babies comparing poractant alfa in two doses of 200 mg/kg and 100 mg/kg with beractant 100 mg/kg for treatment of RDS, there was more rapid improvement in oxygenation in both groups of babies treated with poractant alfa, less need for repeat dosing with the higher initial dose and improved survival for babies of <32 weeks’ gestation who were treated with 200 mg/kg of poractant alfa.41 A meta-analysis of all five randomised trials comparing these surfactants has also shown a mortality difference (OR 0.35; 95% CI 0.13 to 0.92).37 However, the numbers of babies studied is still relatively small and the recent American Academy of Pediatrics recommendations state: “It is unclear whether significant differences in clinical outcomes exist among the available [animal-derived surfactant] products”.6

OTHER INDICATIONS FOR SURFACTANT

Surfactant is used to manage neonatal lung disease other than RDS even though there is a lack of evidence of substantive benefit from randomised trials. Other indications include meconium aspiration syndrome, persistent pulmonary hypertension, pulmonary haemorrhage, pneumonia and delivery of drugs directly to the lung.

Meconium aspiration can lead to severe respiratory failure and this may be partly related to secondary surfactant inactivation. Four randomised trials of surfactant therapy in meconium aspiration syndrome have been included in a systematic review that found improved oxygenation and a reduction in the need for extracorporeal membrane oxygenation (ECMO) (RR 0.64; 95% CI 0.46 to 0.91).42 The studies in this Cochrane meta-analysis used a 6-hourly dosing regimen of natural bovine surfactant for up to four doses. More recent studies have examined dilute surfactant lavage as a means of removing meconium particles from the lungs and future studies may be directed at comparing this method with standard bolus dosing.43 Surfactant has also been used in the setting of congenital diaphragmatic hernia, although this practice is not supported by evidence from randomised trials. However, for babies on ECMO concomitant surfactant therapy shortens the duration of cannulation.44

Surfactant has also been used to treat massive pulmonary haemorrhage, the rationale being that blood is known to inhibit surfactant function. The evidence for benefit of this strategy mainly comes from observational studies as randomised controlled trials are difficult to perform because of the unpredictable nature of the problem.45 Observational studies also show a beneficial effect from surfactant for babies born with group B streptococcal pneumonia in terms of improving oxygenation, at least in the short term.46 Natural surfactants also improve oxygenation in older children who need mechanical ventilation because of acute RDS secondary to pneumonia47 and the first randomised trials have recently confirmed these potential benefits.48

Surfactants may potentially be used for the delivery of drugs directly into the lung, including antibiotics, anti-inflammatory agents and bronchodilators. Recently in a pilot study, 116 very low birth weight infants with severe RDS were randomised to receive either 8-hourly beractant 100 mg/kg or the beractant with added budesonide 0.25 mg/kg. The combined outcome of death/BPD was significantly reduced in the budesonide group in this small study.49 This promising technique deserves further study in a larger trial designed to see if this apparent benefit is real.

THE FUTURE

Natural surfactants will continue to be used for the indications discussed above. For babies with RDS, we will hopefully get a clearer picture of when to treat babies on CPAP with surfactant and at what gestational age routine intubation and surfactant prophylaxis should be used. New generation synthetic surfactants may eventually replace natural surfactants, especially if they can be produced more cheaply. Alternative methods of surfactant administration that forgo the need for intubation may be developed. For example, surfactant has been recently reported to have been successfully administered directly into the trachea at laryngoscopy using a fine feeding catheter,50 by pharyngeal deposition51 or through a laryngeal mask.52 Further studies of direct steroid delivery to the lung of the very preterm infant to reduce BPD are likely to take place and ultimately this may become a standard of care to minimise lung inflammation. Further studies also will be needed to widen the indications for surfactant therapy beyond the prevention and treatment of neonatal RDS, which has been one of the great successes in neonatal medicine in the past 25 years.39