 |
|
Evidence Based Medicne Indian Pediatrics 2002; 39:347-361 |
||||||||||||||||||||||||||||||||||||||||||||
|
Steroids in Perinatology |
||||||||||||||||||||||||||||||||||||||||||||
|
The use of antenatal corticosteroids to accelerate fetal lung maturity has been a revolutionary intervention in perinatology. This low-cost, low-technology preventive strategy has been responsible for saving many newborns and has helped to reduce morbidity beyond doubt. However, many controversies continue to dog the use of antenatal steroids. This article discusses these issues in the light of available evidence in medical literature.
Infants born preterm are at high risk of developing respiratory distress syndrome (RDS) which has its own attendant mortality and morbidity. Therefore, the prevention of RDS in those born preterm makes sound clinical sense. Liggins, in the late 1960s, noted that the functional maturation of the lungs of lambs born preterm was enchanced by maternal glucorticoid administration(1). Similar evidence came from studies on rabbits(2). Autopsies done on infants dying within 72 hours of birth showed that the adrenals were 19% smaller in those dying of hyaline membrane disease(3). By this time, evidence was also available that the human placenta is permeable to glucocorticoids and that maternal administration would lead to increased fetal levels of glucorticoid activity. This prompted Liggins and Howie to try the prevention of RDS in those born preterm by maternal betamethasone administration in the first clinical trial in 1969(1). It is now known that glucocorticoid receptors appear in the lungs as early as 12 weeks of gestation(4) and that the synthesis and storage of surfactant may occur before the full structural development of the alveolar system(5). Fetal lung maturity and cortisol levels have also been shown to correlate(6). All this forms a sound basis for the use of antenatal steroids to accelerate fetal lung maturity. However, clinching evidence comes from the numerous clinical trials that have, beyond doubt, established the value of antenatal steroids.
One study had shown a reduction in the incidence of Chronic Lung Disease from 35% to 25% with the use of antenatal steroids(23) but this has not been replicated by other workers. The National Institute of Health (USA), in its consensus statement on antenatal cortico-steroids has provided the quality of evidence and strength of recommendation for the use of antenatal steroids(6). This is shown in Table I.
Table I__ Evidence of Efficacy of Corticosteroids and Strength of Recommendation According to Delivery Interval, Gestational Age, Status of Membranes and Neonatal Outcome
Q 3. Which drug is to be used?
The initial trials used arbitarily selected doses of betamethasone and dexamethasone. The current dose recommendation is(6):
At these doses, it has been found that delivery to the fetus is optimal(6,24-26) and ensures 75% occupancy of the available fetal corticosteroid receptors(6). Higher or more frequent doses do not, in any way, enchance benefits; on the other hand, the likelihood of adverse effects increases.
The maximum benefit from antenatal steroids in the form of reduction in RDS and mortality is seen in neonates delivered between 29 and 34 weeks(6). For neonates delivered between 24 and 28 weeks of gestation, though the incidence of RDS may not be markedly reduced, its severity is reduced and the incidence of IVH is also reduced(6). Even in this group, there is a trend towards the same benefits as in the older group(30) and steroids positively influence survival(31). Beyond 34 weeks of gestation, the risk of RDS and IVH is low(6,8,10) hence antenatal steroids are not to be used. The current recommendation thus is that antenatal steroids should be used for all fetuses between 24 and 34 weeks of gestation and at risk of preterm delivery.
After one course (total of 24 mg of Betamethasone or Dexamethasone, given over 48 hours) of antenatal steroid, the maximum benefits are seen if the fetus is delivered 24 hours after and within 7 days of the last dose(6). Those who deliver within 24 hours of the last dose or even before the full course can be completed, do share some marginal benefits (6). It has also been shown that improved neonatal outcomes continue even beyond 7 days of the last dose and that the risk of RDS substantially decreases in women who deliver more than 7 days after the onset of premature labor, irrespective of whether they received antenatal steroids or not(32). Therefore, use of antenatal steroids in a single course ensures definite benefit to a fetus at risk of preterm delivery.
With the evidence available now the answer is an emphatic NO - repeat courses are not necessary. Doren and co-workers(10) in their trial from 1975 to 1978 with betamethasone, were the first to note that benefits of therapy persisted beyound 7 days of the last course. They recommended only a single course. Though many studies have used repeat weekly courses of steroids till 34 weeks gestation or delivery, there is no scientific basis for this. In fact, the 1994 NIH Consensus statement(6) listed the use of repeat courses as an area for further research. Animal studies have shown delayed myelination and decreased growth of all areas of the fetal brain, especially the hippocampus, following repeated antenatal steroid use(33,35). Critics of these studies point out that proportionately higher doses of antenatal steroids were used in these animal studies. A recent retrospective study on 713 infants who had received 1 to 12 courses of antenatal betamethasone or dexamethasone has revealed a reduction in the head circumference at birth by 0.46 ± 0.19 cm in the group having received more than one course of steroids(36). A prospective study of 453 patients of whom 186 received more than 1 course of antental steroids has shown an increased risk of early neonatal sepsis (OR 5.00; 95% CI 1.3-2.32) and neonatal death (OR 2.92; 95% CI 1.3 - 6.9). The risk of chorioamnionitis and maternal endometritis also increased with multiple courses of steroids. There was no difference in the incidence of RDS or severe IVH (Grade III/IV) between the single and multiple courses groups(37). Other workers have reported transient hypertropic cardiomyopathy, suppression of fetal pituitary-adrenal axis, and fetal growth restriction without any increase in benefits(38,39). In view of the fact that the benefit of a single course lasts beyond 7 days and the fact that the incidence of RDS decreases drastically if delivery does not occur with 7 days of onset of threatened labor, and given the evidence of adverse effects of multiple courses of antenatal steroids, it is strongly recommended that only a single course be used.
Initial workers had doubts about the use of antenatal steroids in women with PPROM because of the risk of infection and the perceived protection against RDS offered by PPROM itself. However, controlled trials soon proved that PPROM did not protect against RDS and that antenatal steroids did not increase infection rates in PPROM(15). Though the risk of infection may be theoretically increased in PPROM, the benefits of antenatal steroids far outweigh the minimal increase in risk of infection(6). One small study on 94 infants reported a decreased incidence of RDS in females delivered to mothers with PPROM irrespective of steroid use(39) but this was not replicated in other studies. Since then, antenatal steroids given to women with PPROM have been proved to reduce RDS, neonatal mortality, IVH and length of hospital stay without any increase in maternal or neonatal sepsis(40,42). An increase in neonatal sepsis and chorioamnionitis has recently been reported with multiple courses of steroids, even in those with intact membranes(37) while one study reported an increased risk of infection in women with PPROM who had a cervical suture in situ(40). Chorioamnionitis remains a contraindication to the use of antenatal steroids. With available evidence, the use of a single course of antenatal steroids in the presence of PPROM without chorioamnionitis seems highly justified. Antenatal steroids do not interfere with the diagnosis of infection in women with PPROM(6).
The first trial by Liggins and Howie(1) showed increase in fetal demise in women with PIH who were treated with steroids. On closer scrutiny, it turned out that there were more women with severe PIH in the treatment group than in the placebo group - probably a chance error. Subsequent studies have shown no increase in still birth or infection rates in women with PIH treated with antenatal steroids(15,18). In women with diabetes treated with steroids, the monitoring and control of diabetes may become transiently difficult(6). Though data is insufficient to assess benefit of antenatal steroids in maternal PIH, diabetes, hypertension and other illnesses, it definitely does not lead to increased adverse effects. Similarly, multiple pregnancies, fetal hydrops and fetal growth restriction are not per se contraindications for the use of antenatal steroids(6). Therefore, after ensuring adequate treat-ment and control of the maternal illness in question, all fetuses at risk of being delivered preterm should be offered the benefits of a single course of antenatal steroids.
Remember, prevention is always better than cure!! In a developing country like ours, the cost of surfactant still remains very high and unaffordable for most patients. In sharp contrast, antenatal steroids are very cheap as well as effective in preventing RDS. Moreover, surfactant replacement does not have other benefits like reduction in IVH and PDA(6). Lastly, antenatal steroids act additively with the use of postnatal surfactant to decrease the incidence and severity of RDS, to decrease mortality and IVH(6,43). The availability of surfactant replacement therapy should not preclude the use of antenatal steroids.
Since maximum benefits of antenatal steroids are seen in neonates delivered 24 hours after the completion of a course(6), it makes sound sense to try and postpone delivery till 24 hours after completion of a course of steroids. Unless contraindicated, tocolysis should be offered to women in preterm labor to allow fetuses to get maximum benefit of the antenatal steroids. One study on use of betamethasone with ritodrine tocolysis showed a reduction in RDS, ventilator dependency and in those who survived for >48 hours, a reduction in PDA(18). Beta-mimetics, commonly used for tocolysis, increase the risk of IVH in the neonate and this is countered by antenatal steroids(18). Other than this, there is a theoretical risk of maternal pulmonary edema when tocolytics are used in conjunction with antenatal steroids(6). Tocolytics cannot be used in conditions where delay in delivery may worsen the maternal medical status. In all other situations, it would be beneficial to offer tocolysis and delay preterm labor for a period in which the fetus gets maximum benefits of antenatal steroids.
The first trial with dexamethasone reported an increased trend for neonatal infection(7) while early hypoglycemia was reported by Papageorgiou’s group(8). The meta-analysis of all these trials did not reveal any significant increase in neonatal infection(15) and this has been reflected in the NIH consensus statement(6). Fetal cortisol levels have been shown to drop within 6 hours of the first betamethasone dose and return to normal by 7 days(44). There is no detrimental effect on mineralocorticoid sufficiency or response of the neonatal pituitary-adrenocortical axis to stress after birth(44,45). Also reported, though clinically insignificant, is lower cord growth hormone levels(44) and increased cord prealbumin levels(46). Overall, there seem to be no adverse effects of antenatal steroids on the fetal- neonatal endocrine system. The first case of neonatal leukocytosis following antenatal betamethasone was reported in 1978(47). Subsequently, one neonate in 1981(48) and five in 1982(49) were reported to have transient leukocytosis. Interestingly, all these patients were born after PPROM, had transient, mild RDS after birth and all but one were female. The leukocytosis was of the order of 60,000 to 1,00,000/cu mm, consisted predominantly of polymorpho-nuclear cells and their precursors and the counts returned to normal by day 6 to 7 of life without any ill effects. The Collaborative Group on Antenatal Steroid Therapy from Maryland, USA, published their findings in 1981 and this included a Prechtl examination done at 40 weeks corrected age(12). The results were encouraging in that there was a trend (though not statistically significant) towards better performance by the group treated with antenatal steroids. One study reported an increase in VLDL and LDL cholesterol in term infants and increased HDL cholesterol in preterm in- fants whose mothers were treated with a combination of betamethasone, phenobarbi-tone and ritodrine tocolysis(50). This increase was not seen in infants of mothers treated with phenobarbitone alone. No other study has reported similar findings. On the whole, it is now amply clear that the use of antenatal steroids is devoid of any short-term adverse effects on the neonate.
Morrison et al.(51) reported one of the earliest follow-ups of infants of mothers treated with antenatal steroids. There were no adverse effects noted at 3½ years of age but the follow up was not complete and did not include neurological testing. A four-year assessment of the survivors of the first trial(1) was reported by MacArthur, Howie and others. Despite the betamethasone treated group having more prematurely delivered subjects, they found no difference in cognitive and psychosocial development(52). A 3-year follow-up by the Collaborative Group on Antenatal Steroid Therapy included 406 of the original infants and did not find any difference in development between the treated and control groups(53). Dutch workers followed children for their psychological and physical development at 1-12 years age. They did not find any differences between the two groups(54,55). Doyle et al. reported normal 5-years outcomes of Extremely Low Birth Weight (<1000 g) infants who had received antenatal steroids(56). Rennie and co-workers from the UK reported better long-term out-comes for Very Low Birth Weight (<1500 g) infants treated with antenatal steroids(17). School progress and cognitive development of 6-year old children were assessed in 82.2% of survivors of Liggins and Howie’s original trial. This study too did not find any differences between the treated and untreated groups(57). Concerns about long term neurological adverse effects of antenatal steroids arose from early animal studies(6,10) but the doses used were 10 to 30 times the equivalent doses used in human trials. A retrospective study has associated the use of repeated courses of antenatal steroids with decreased head circumference at birth(36). Postnatal trials of dexamethasone while brain maturity is still incomplete have shown increase in disability in the treated group(27). Thus, current evidence shows us that the judicious use of a single course of beta-methasone is, in all probability, not associated with any long-term adverse neurodevelopment outcome.
Apart from the evidence that repeated course of antenatal steroids increase the incidence of chorioamnionitis and maternal endometriosis(37) there is no significant adverse effect on the mother. Pulmonary edema is a theoretical risk in mothers who receive tocolytics with antenatal steroids and is more common in the presence of maternal infection, fluid overload and multiple gestation(6). Antenatal steroids do not increase the risk of maternal infection, irrespective of the status of the mem-branes(15). A single course of antenatal steroid is safe and devoid of any maternal adverse effects.
Antenatal Thyrotropin-Releasing Hor-mone (TRH) is another agent that has been used in clinical trials and is mentioned in the NIH consensus statement(6). The Australian Collaborative Trial of Antenatal Thyrotropin-Releasing Hormone (ACTOBAT) study, where TRH was used along with antenatal betamethasone, found increased risk of RDS and increased ventilatory needs in the treated group. More worrisome is the fact that at 12 months follow up, treatment with TRH was associated with global delay of milestones as well risk of severe impairment of develop-ment(58). The use of antenatal TRH, therefore, is not recommended at present. Postnatal Steroids In contrast to antenatal steroids, the use of postnatal steroids to prevent or modify Chronic Lung Disease (CLD) or Broncho-pulmonary Dysplasia (BPD) is riddled with controversies. Evidence of the benefits of postnatal steroids, their adverse effects and current recommendations are discussed.
Initial investigators used early postnatal steroids to try and modify the course of RDS. The rationale behind this was evidence of acceleration of lung maturation seen with antenatal steroids and in animal experi-ments(59). This practice, however, did not work and was soon abandoned. With time, more numbers of smaller infants with RDS were salvaged. BPD then began to emerge as a major morbidity in these survivors. It was postulated that postnatal steroids would act to suppress or abolish the inflammatory responses that lead to BPD. Thus, earlier trials used dexamethasone in the third week of life to modify BPD. In the mid-1990s, evidence of higher levels of inflammatory cytokines and inflammatory cells in the lung lavage fluid of infants with a prediliction for CLD was forthcoming. Salient findings of these studies(60-62) included the following: (a) Inflammatory responses to microbes and non-specific injury (barotrauma/hyperoxia) were similar; (b) Early rise in mediators of inflammation in lung lavage fluids (at 10 days age) could predict infants who were at risk of CLD/BPD; (c) Increased pulmonary microvascular permeability was a part of CLD; and (d) An imbalance in the protease-antiprotease levels in the lungs of those who develop BPD. With evidence of the early onset of in-flammation leading on to CLD/BPD, steroids started being administered early (<14 days age) to prevent CLD/BPD. These facts concerning the patho-physiology of CLD/BPD do seem to provide a scientific basis for the use of postnatal steroids to prevent or attenuate CLD/BPD in high-risk infants.
One of the first controlled trials by Avery and his team in 1984 was conducted on 16 ventilator dependent infants who were given dexamethasone in the fourth week of life. Treated infants showed substantial short-term improvement in lung function and permitted early weaning off the ventilator(63). By the late 1990s, a number of trials had been conducted with varying doses of dexamethasone started at different ages and continued for varied durations. A meta-analysis of ten such trials, in which dexa-methasone was started within the first 14 days, showed a significant reduction in the risk of CLD (both at 28 days age and 36 weeks post-menstrual age). Subgroup analysis revealed a significant reduction in mortality risk in those started on dexamethasone between 7 and 14 days of age(64). Another meta-analysis with stricter inclusion criteria for trials concluded that early (<15 days) dexamethasone therapy for at risk infants reduced risk of CLD without significant short term adverse effects(65). A Cochrane meta-analysis on early post-natal (<96 hours) steroid use did find benefits in the form of earlier extubation, decreased risk of CLD at 28 days and 36 weeks, decreased death or CLD at 28 days and reduction in risk of PDA. Significant adverse effects were gastrointestinal bleeding and spontaneous intestinal perforation. The risks of hyper-tension and hyperglycemia increased. The reviewers concluded that the benefits of early postnatal steroids may not outweigh the risks(66). A similar meta-analysis on steroid use beyond three weeks of age for CLD revealed reduction in extubation failure, reduced use of dexamethasone rescue therapy later, reduction in CLD at 36 weeks and home oxygen therapy. Mortality risk was not affected. Hyper-glycemia, hypertension and glucosuria were short-term adverse effects while abnormal neurological examination and increased risk of cerebral palsy were noted in the long term. The meta-analysis concludes that steroids should be reserved for infants who cannot be weaned from the ventilator and the lowest dose for the shortest duration is to be used(67).
Data on effects of postnatal inhaled steroids is limited. Only four clinical trials could be included in the Cochrane review on use of inhaled steroids to modify existing CLD. The review revealed facilitation of extubation by inhaled steroids without any increase in sepsis rates(68). Review of 5 trials of early (<15 days) treatment with inhaled steroids did not show any reduction in CLD(69) while the recent OSECT (Open Study of Early Corticosteroid Treatment) study did not find any significant reduction of CLD with early (<3 days) or late (>15 days) therapy with either systemic dexamethasone or inhaled budesonide(70). Thus, there is no evidence of definite benefit in the use of inhaled steroids for CLD.
The use of dexamethasone for facilitating extubation has been reviewed and the verdict is that its use should be reserved for neonates who have been repeatedly intubated or have remained intubated for prolonged periods(71). The benefits of postnatal systemic steroids do not outweigh its risks and should be used very selectively, in low doses for the shortest duration, in infants who have failed earlier extubation attempts. A low dose of dexa-methasone of 0.125 mg/kg intravenously, given 12 hourly for a total of 6 doses has been found to facilitate weaning from the ventilator(72).
A recent study using low dose dexamethasone for 10 days, found hyper-tension and increased spontaneous gastro-intestinal perforations without any reduction in CLD in a group of treated infants <1000 gram. In fact, this study was terminated because of the high rates of adverse effects(73). Dexa-methasone is also known to reduce weight gain during therapy and lead to decreased head growth(73-74). It also increases the risk of systemic infection(75). The more worrisome adverse effects of postnatal dexamethasone relate to the central nervous system. Yeh et al.(76) followed up infants who had received early (<12 hours) dexamethasone. At 2 years corrected age, the treated group had significantly increased neuromotor dysfunction and poorer per-formance on the psychomotor development index (PDI) of the Bayley Scales of Infant Development. Other workers have found a higher incidence of cerebral palsy in the treated group as compared to the placebo group (49% versus 15%) in a trial of early three-days course of dexamethasone therapy(77). A recent study analyzed 3-D MRI images of infants treated with postnatal dexa-methasone has found significantly impaired growth of cerebral gray matter and a 30% reduction in the total cerebral tissue volume as compared to controls(78). Thus, postnatal steroid use is not free of significant short term as well as long term adverse effects.
The most commonly used drug is dexamethasone though different workers have used different doses and given therapy for varied durations. Some of the commonly used regimes are as follows:
It is once again emphasized that the benefits of use of postnatal steroids do not outweigh the risks and hence should be used with discretion.
Postnatal steroids have been used in resistant hypoglycemia, congenital adrenal hyperplasia, and sepsis shock syndrome with sclerema. These conditions are infrequent and a detailed discussion on these is beyond the scope of this article. Readers are requested to refer to standard textbooks of neonatology for more details. Contributors: SN reviewed and drafted the paper. AKD provided overall framework and concept; he will act as the guarantor of the article. Funding: None. Competing interests: None stated.
| ||||||||||||||||||||||||||||||||||||||||||||
| References | ||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||
