Correspondence
Antenatal distress in infants at 34 weeks’ several of the concerns expressed in corticosteroids to gestation. This statement is based on the Comment by Azad and Costello a subgroup analysis from a systematic will be addressed. 3 reduce preterm deaths review. However, the same review We declare that we have no competing interests. presents data showing a decreased in low-income settings Copyright © Althabe et al. Open Access article risk of respiratory distress syndrome distributed under the terms of CC BY-NC-ND. In their Comment (April issue),1 in infants with first dose of cortico- Fernando Althabe, José M Belizán, Kishwar Azad and Anthony Costello steroids administered to mothers Pierre Buekens, *Elizabeth M McClure, raise questions that should be answered at 33–35 weeks’ gestation (relative Marion Koso-thomas, on behalf of the before antenatal corticosteroid treat- risk [RR] 0·53, 95% CI 0·31–0·91), NICHD’s Global Network for Women’s ment is scaled up to reduce pre term and a non-signifi cant decrease in the and Children’s Health Research ACT deaths in low-income countries. We risk of respiratory distress in infants Trial Steering Committee share their concerns about the unknown (0·61, 0·11–3·26) with first dose at [email protected] overall effect of this treatment on 35–37 weeks’ gestation. The fi ndings Institute for Clinical Eff ectiveness and Health Policy, mortality and potential safety issues in suggest a reduction in respiratory Buenos Aires, Argentina (FA, JMB); Tulane School of the mother. To answer these questions, distress syndrome is present according Public Health and Tropical Medicine, New Orleans, we have initiated the Antenatal to gestational age at fi rst delivery of LA, USA (PB); RTI International, 3040 Cornwallis Drive, Durham, NC, 27709, USA (EMM); Eunice 2 3 Corticosteroids Trial to assess whether corticosteroids. Kennedy Shriver National Institute of Child Health or not a multifaceted intervention Prevention of respiratory distress and Human Development, Rockville MD, USA (MK-T) to increase the use of antenatal syndrome in infants born at 1 Azad K, Costello A. Extreme caution is needed corticosteroids reduces neonatal 33–36 weeks’ gestation without access before scale-up of antenatal corticosteroids to reduce preterm deaths in low-income settings. mortality at 28 days of age, and to specialised high-quality level 2 care Lancet Glob Health 2014; 2: e191–92. maternal morbidity due to infections. might create a substantial health-care 2 Althabe F, Belizán JM, Mazzoni A, et al. Antenatal corticosteroids trial in preterm births to increase Enrolment has been completed and burden in low-income countries. The neonatal survival in developing countries: study data from more than 90 000 births have Antenatal Corticosteroids Trial2 will protocol. Reprod Health 2012; 9: 22. been collected. assess the administration of steroids 3 Roberts D, Dalziel SR. Antenatal corticosteroids for accelerating fetal lung maturation for We disagree with Azad and Costello’s to mothers up to 36 weeks’ gestation. women at risk of preterm birth. comment about the eff ect of antenatal Data from this trial will be available in Cochrane Database Syst Rev 2006; 3: CD004454. corticosteroid treatment on respiratory the second half of 2014. We hope that
www.thelancet.com/lancetgh Vol 2 August 2014 e444 Correspondence
Antenatal done as part of a trial, and focuses only corticosteroids are likely to have a corticosteroids to on quality improvement, we noted greater eff ect in the absence of level 2 no increase in the rate of maternal or care, not a lesser eff ect.4 The diff erence reduce preterm deaths neonatal infections. between low-income and high-income in low-income settings Antenatal corticosteroids induce fetal settings is not biology, but an increased lung maturation through the same burden of disease and reduced access to The Comment1 by Kishwar Azad and biological mechanism in low-income even basic health care. Our experience Anthony Costello opposing scale- settings as in high-income settings in Malawi off ers a powerful example for up of antenatal corticosteroids and reduce the need for neonatal generalising this standard of care to the misdirects the discussion of this mechanical ventilation.3 Although regions where it will save the most lives topic towards speculation about antenatal corticosteroids might not be and also reduce neonatal disability. differences in low-income settings. a so-called magic bullet as a standalone I declare no competing interests. Our experience in Malawi provides vertical intervention, no biological Copyright © Kaliti. Open Access article distributed a concrete example of the rapid basis exists to presume that babies under the terms of CC BY-NC-ND. scaling up of antenatal corticosteroid born preterm in resource-poor settings Stephen Kaliti treatment with dexamethasone. will succumb to respiratory distress [email protected] Malawi has the highest estimated syndrome any more than do those in 2 Bwaila Maternity Hospital, Obstetrics and preterm birth rate worldwide. In resource-rich countries. I support the Gynecology, Malanga Langa Road, Lilongwe, Malawi Bwaila Maternity Hospital, Lilongwe, existing recommendation of a single- 1 Azad K, Costello A. Extreme caution is needed that has more than 15 000 deliveries course of antenatal corticosteroids to before scale-up of antenatal corticosteroids to annually with more than 2900 preterm, mothers at high risk of preterm birth reduce preterm deaths in low-income settings. Lancet Glob Health 2014; 2: e191–92. we increased targeted coverage of between 24 weeks and 33 weeks’ 2 Blencowe H, Cousens S, Oestergaard M, et al. antenatal corticosteroids from 8% to gestation, but question Azad and National, regional and worldwide estimates of 80% in 16 weeks in women at risk of Costello’s unrealistic prerequisite for preterm birth rates in the year 2010 with time trends for selected countries since 1990: preterm delivery from 24 to 34 weeks’ round-the-clock access to level 2 care in a systematic analysis and implications. Lancet gestation. After this pilot study, we a low-income setting. 2012; 9832: 2162–72. 3 Roberts D, Dalziel S. Antenatal corticosteroids for began programmes in three other Low-income settings, which have the accelerating fetal lung maturation for women at hospitals, reaching 59–83% coverage highest burden of preterm neonatal risk of preterm birth. Cochrane Database Syst Rev from a baseline of 1–6% within 6 weeks. deaths, urgently need proven benefi cial 2006, 3: CD00004454. 4 Mwansa-Kambafwile J, Cousens S, Hansen T, This intervention has thus far been interventions, not the assessment Lawn JE. Antenatal steroids in preterm labour associated with a drop in preterm of therapeutic efficacy on the basis for the prevention of neonatal deaths due to complications of preterm birth. Int J Epidemiol neonatal mortality contribution of resource profi ling that could delay 2010; 39: 1122–33. from 60% to 24% at 0–6 days of age. treatment. Contrary to Azad and Although this intervention was not Costello’s speculation,1 antenatal
www.thelancet.com/lancetgh Vol 2 August 2014 e445 Correspondence
Antenatal of prematurity so their concerns with of antenatal corticosteroids. We support corticosteroids to respect to perinatal death or disability the call for more research, especially are hard to justify. Repeat antenatal on how to reach the poorest women reduce preterm deaths corticosteroids have been linked to and how to increase long-term health in low-income settings learning disabilities compared with a for both women and their babies. In single dose,4 and late-onset metabolic the meantime, the evidence strongly Kishwar Azad and Anthony Costello1 syndrome might also be a risk.5 With supports giving a single, short course suggest the use of extreme caution in respect to maternal outcomes there of corticosteroids to women at risk of scaling up of antenatal corticosteroid is no robust evidence of increased preterm birth in hospitals everywhere, treatment in low-income settings. infections.6 Because preterm deaths are not just in high-income countries. They raise three important questions now the leading cause of child deaths at Prof Joy E Lawn is co-lead of ACS Technical with respect to the effi cacy, safety, 1 million per year, the balance lies in the Reference Team for UN Commission on Life Saving and the appropriate gestational age direction of reducing mortality rather Commodities and wrote this Correspondence on behalf of the group. at which to give corticosteroids to than the unknown risks of less severe Copyright © Lawn et al. Open Access article patients in low-income countries. outcomes. distributed under the terms of CC BY-NC-ND. Firstly, in terms of efficacy, there Thirdly, although the proven benefi t *Joy E Lawn, Joel Segre, Pierre Barker, is high-quality evidence on the of antenatal corticosteroids is when Jeff rey Smith, Irene De La Torre, benefits of antenatal corticosteroids they are administered to patients William Stones for lung maturation in utero. A large at 28–33 weeks’ gestation, this [email protected] decrease in neonatal mortality was gestational age band is partly due London School Hygiene and Tropical Medicine, reported in trials in four middle- to enrolment criteria in the original University College London, UK and Save the Children, income countries, including those in trials.2,3 The gestational-age limit for South Africa (JEL); Consultant to Bill & Melinda Gates Africa and the Middle East (relative antenatal corticosteroids in high- Foundation, Seattle, WA, USA (JS); Institute for Healthcare Improvement, Cambridge, MA, USA and risk [RR] 0·47, 95% CI 0·35–0·64), income countries has been extended Gillings School of Global Public Health, University of compared with 14 studies in high- with guidelines supporting use at North Carolina Chapel Hill, NC, USA (PB); Jhpiego, income countries (0·79, 0·65–0·96).2,3 less than 26 weeks’ gestation. More Baltimore MD, USA (JS); University of Puerto Rico, Antenatal corticosteroids actually than 85% of preterm infants are born San Juan, Puerto Rico and International Confederation of Midwives, The Hague, Netherlands 7 reduced the need for level 2 care, at least 32 weeks’ gestation, and (IDLT); and University of St Andrews, St Andrews, UK including mechanical ventilation or although few have major preterm birth and FIGO Committee on Safe Motherhood and continuous positive airway pressure complications, this amounts to a large Newborn Health, London, UK (WS) in four studies (0·69, 0·53–0·90) proportion of infants potentially at 1 Azad K, Costello A. Extreme caution is needed before scale-up of antenatal corticosteroids to and intensive care in two studies risk. The upper gestational-age limit for reduce preterm deaths in low-income settings. (0·80, 0·65–0·99) suggesting that, in corticosteroid use is a critical question Lancet Glob Health 2014; 2: e191–92. 2 Roberts D, Dalziel S. Antenatal corticosteroids regions where mechanical ventilation yet to be answered, especially in for accelerating fetal lung maturation for is not available, substantial benefits health-care settings where mechanical women at risk of preterm birth. Cochrane could be expected.2,3 We agree that ventilators are not widely available Database Syst Rev 2006; 4: CD00004454. 3 Mwansa-Kambafwile J, Cousens S, Hansen T, more research is needed but in and antenatal corticosteroids are Lawn JE. Antenatal steroids in preterm labour view of the biological basis for the more likely to be life-saving. WHO is for the prevention of neonatal deaths due to complications of preterm birth. Int J Epidemiol effect of antenatal corticosteroids presently review ing the recommended 2010; 39: 1122–33. on infant mortality, it is extremely upper and lower gestational-age 4 Crowther CA, Harding JE. Repeat doses of unlikely, statistically, that antenatal cutoffs for antenatal corticosteroid prenatal corticosteroids for women at risk of preterm birth for preventing neonatal corticosteroids would be shown not to treatment. When gestational-age of respiratory disease. Cochrane Database Syst Rev work in African or Asian babies. an infant is unknown or is imprecisely 2007; 3: CD003935. Secondly, we agree with Azad known, the balance of risks needs to 5 Seckl JR1, Cleasby M, Nyirenda MJ. Glucocorticoids, 11beta-hydroxysteroid and Costello that potential harm to be considered and in high mortality dehydrogenase, and fetal programming. the patient is always a critical issue. settings the balance will be in favour of Kidney Int 2000; 57: 1412-7. 6 Crowther C, Brown J. Comment on: extreme However, a one-off course of antenatal treatment. caution is needed before scale-up of antenatal corticosteroids (<48 h) poses a High-income countries have at corticosteroids to reduce preterm deaths in very low risk of adverse effects. The least 90% coverage of antenatal low-income settings. Lancet Glob Health 2014; 2: e447 2 Cochrane systematic review discussed corticosteroids, with most women 7 Blencowe H, Cousens S, Oestergaard MZ, et al. by Azad and Costello shows antenatal in preterm labour being treated, and National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time corticosteroids are associated with clinicians would be sued for non-use. trends since 1990 for selected countries: major reductions in, death, severe Yet countries with the highest rates of a systematic analysis and implications. Lancet disability and lower rates of retinopathy preterm births have negligible coverage 2012; 379: 2162–72.
www.thelancet.com/lancetgh Vol 2 August 2014 e446 Correspondence
Antenatal (0·57, 0·17–1·89; 204 women) and one present recommendations, the major corticosteroids to trial from Jordon (4·19, 0·94–18·68; safety concern surrounding the use 139 women). Incidence of maternal of antenatal corticosteroids is repeat reduce preterm deaths postnatal fever did not differ in doses.5 in low-income settings two trials, the US Collaborative trial CC is an author of two of the Cochrane systematic (0·93, 0·56–1·53; 682 women) and reviews cited in this Correspondence and is principal In their Comment1 on the use of the Dexiprom trial in South Africa investigator for the A*STEROID randomised controlled trial (ACTRN12608000631303). antenatal corticosteroids to reduce (1·00, CI 0·36–2·75; 204 women). Most Copyright © Crowther et al. Open Access article preterm infant deaths, Kishwar Azad reassuringly, no signifi cant diff erence distributed under the terms of CC BY. and Anthony Costello advise “extreme was reported in the incidence of caution” before scale-up in low-income chorioamnionitis in four trials of *Caroline Crowther, Julie Brown settings. They emphasise maternal dexamethasone (1·35, 0·89–2·05; [email protected] sepsis as a concern but cite only one 575 women) or postnatal fever Liggins Institute, University of Auckland, trial in which dexamethasone resulted in two trials of dexamethasone Auckland 1023, Grafton, New Zealand in a significant increase in fever (0·94, 0·60–1·47; 886 women).2 No 1 Azad K, Costello A. Extreme caution is needed before scale-up of antenatal corticosteroids to that required antibiotic treatment trials of dexamethasone reported on reduce preterm deaths in low-income settings. compared with controls (relative maternal intrapartum fever when Lancet Glob Health 2014; 2: e191–92. risk [RR] 2·05, 95% CI 1·14–3·69; antibiotics were given. 2 Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lung maturation for 1,2 118 women). We suggest that Trials of betamethasone versus women at risk of preterm birth. Cochrane this fi nding alone does not refl ect a dexamethasone in accelerating fetal Database Syst Rev 2006; 3: CD00004454. 3 Brownfoot FC, Crowther CA, Middleton P. balanced assessment of the paucity of lung maturation have not reported Diff erent corticosteroids and regimens for evidence available.2 on maternal infectious outcomes.3 accelerating fetal lung maturation for women at In a systematic review of antenatal We are currently undertaking a large- risk of preterm birth. Cochrane Database Syst Rev 2008, 4: CD00006764. corticosteroid treatment to accelerate scale trial to compare the efficacy 4 Crowther CA, Harding JE, Middleton PF, et al. fetal lung maturation, only four of intramuscular dexamethasone Australasian randomised trial to evaluate the role of maternal intramuscular dexamethasone of 21 randomised controlled trials versus betamethasone in reducing versus betamethasone prior to preterm birth to report on puerperal sepsis outcomes childhood neurosensory disability, increase survival free of childhood neurosensory for dexamethasone versus no ante- with maternal infection as a disability (A*STEROID): study protocol. BMC Pregnancy Childbirth 2013; 13: 104. 4 natal corticosteroids and these secondary outcome. Currently, no 5 Crowther CA, McKinlay CJD, Middleton P, show moderate heterogeneity published data suggest a major risk Harding JE. Repeat doses of prenatal corticosteroids for women at risk of preterm (I² 38%) (RR 1·74, 95% CI 1·04–2·89; of maternal infection with the use birth for improving neonatal health 536 women).2 Only two trials were of antenatal corticosteroids and outcomes. Cochrane Database Syst Rev 2011, in low-income to middle-income none are available to allow confi dent 6: CD003935. countries and had very diff erent results: assertion that dexamethasone the Dexiprom trial from South Africa increases the risk. According to
www.thelancet.com/lancetgh Vol 2 August 2014 e447 Correspondence
Antenatal there is no evidence of a similar eff ect have a greater eff ect in the absence of corticosteroids to in populations in high-income regions level 2 care, not a lesser eff ect”, but the in the USA or Europe.1,2 evidence to support this statement is reduce preterm deaths Our concern arose because of weak. Any policy to extend antenatal in low-income settings reports implying that antenatal corticosteroid delivery to mothers corticosteroids could be scaled up as through community health-care Authors’ reply a vertical treatment administered to workers in regions where access to We agree with many of the women with signs of preterm labour good-quality specialised care is not correspondents’ points. First, we by community health-care workers, or available (and where the assessment concur that antenatal corticosteroid at outreach clinics without specialised of gestational age and duration treatment can reduce respiratory level-2 health-care facilities. Two of pregnancy is often unreliable) distress in infants born at less than potential risks from this setting could should be on the basis of randomised 34 weeks’ gestation. Second, we outweigh the benefi ts: the possibility community effectiveness trials. welcome the rapid scale-up of its use that the number needed to treat, to The risks and benefits can then be in hospitals in Malawi, as described save the life of a preterm infant, could measured in the same way that those by Stephen Kaliti, and look forward increase the incidence of serious of dietary vitamin A, chlorhexidine, to the published assessment of sepsis in mothers, and second, death and zinc supplementation have been its effect on mortality in preterm or disability might occur later as a assessed in populations in low-income infants. Third, we agree that more result of suboptimal preterm care. regions.5,6 We urge funders to support research is needed to explore the best Globally, 40 million women deliver these studies. methods for scale-up in hospitals and their babies at home every year and We declare that we have no competing interests. to assess the risks and benefits to many more face formidable economic, Copyright © Azad et al. Open Access article patients through community studies cultural, and geographical barriers distributed under the terms of CC BY. in low-income regions. Particularly, to accessing good-quality maternity Kishwar Azad, *Anthony Costello we anticipate findings of a trial of health care. Many of these women [email protected] antenatal corticosteroid treatment live in low-income regions and rural Perinatal Care Project, Diabetic Association of that is underway in several countries populations in Africa and south Asia Bangladesh, Dhaka, Bangladesh (KA); Institute for and in which 90 000 infants have and endure high levels of malnutrition, Global Health, University College London, been enrolled. especially a lack of micronutrients 30 Guilford Street, London WC1N 1EH, UK (AC) Conversely, we are not in agree- and protein, malaria, anaemia, and 1 WHO Guidelines Approved by the Guidelines Review Committee. Guideline: vitamin A ment with Joy Lawn and colleagues worm infestations that combined supplementation in infants 1–5 months of who suggest that it is, “extremely with the immunosuppressive effect age. Geneva: World Health Organization, 2011. unlikely, statistically, that antenatal of pregnancy or HIV infection might 2 Prost A, Colbourn T, Seward N, et al. Women’s groups practising participatory learning and corticosteroids would be shown not increase their vulnerability to sepsis. action to improve maternal and newborn to work in African or Asian babies”. In the USA, chorioamnionitis aff ects health in resource-limited settings: systematic review and meta-analysis. Lancet 2013; Our point about the potential 9% of pregnancies, but the burden 381: 1736–46. risks associated with antenatal of placental infection is much higher 3 Malloy MH. Chorioamnionitis: epidemiology corticosteroid scale-up was not a in Africa and Asia.3,4 We agree with of newborn management and outcome United States 2008. J Perinatol 2014; published question of efficacy, or ethnicity, Caroline Crowther and Julie Brown online May 1. DOI:10.1038/jp.2014.81. but instead the underlying risks and that a paucity of evidence exists for 4 Chico RM, Mayaud P, Ariti C, Mabey D, benefi ts of antenatal corticosteroids the effect of dexamethasone on Ronsmans C, Chandramohan D. Prevalence of malaria and sexually transmitted and to populations in low-income maternal infection, anywhere, and a reproductive tract infections in pregnancy in regions and diff erent levels of access complete absence of evidence exists sub-Saharan Africa: a systematic review. JAMA 2012 6; 307: 2079–86. to health care. These factors can in low-income settings. Health-care 5 El Arifeen S, Mullany LC, Shah R, et al. strikingly change the benefi t-to-risk workers need to be sure, however, The eff ect of cord cleansing with chlorhexidine ratio of interventions. For example, that antenatal corticosteroids do on neonatal mortality in rural Bangladesh: a community-based, cluster-randomised trial. in low-income regions in south not exacerbate the severity, or the Lancet 2012; 379: 1022–28. Asia and Africa, dietary vitamin A dissemination, of maternal infections 6 Mori R, Ota E, Middleton P, Tobe-Gai R, Mahomed K, Bhutta ZA. Zinc supplementation supplementation in rural populations in these communities. for improving pregnancy and infant outcome. and participatory women’s groups Stephen Kaliti suggests that, Cochrane Database Syst Rev 2012; 7: CD000230. had a large eff ect on child survival, but “antenatal corticosteroids are likely to
www.thelancet.com/lancetgh Vol 2 August 2014 e448 Comment
Extreme caution is needed before scale-up of antenatal corticosteroids to reduce preterm deaths in low-income settings
The great American epidemiologist Bill Silverman facilities”—is speculative. The clamour to roll out a so- Lancet Glob Health 2014 taught us that the road to hell is paved with good called magic bullet treatment must be resisted until Published Online March 13, 2014 intentions. From 1942, during a 12-year period, more three crucial questions have been answered. http://dx.doi.org/10.1016/ than 10 000 infants were blinded through retinopathy The fi rst of these questions is whether antenatal S2214-109X(14)70020-8 of prematurity because paediatricians recommended corticosteroids actually work for mothers and infants a minor increase in the concentration of supplemental at 33 weeks’ gestation or less in poor populations? All oxygen for preterm infants in incubators.1 Later, “untold studies in the systematic review were from hospitals thousands of premature infants succumbed in the fi rst in which infants had access to ”level 2” special care: few days of life because incubator temperatures were 24-h availability of skilled nursing; thermal stability; set slightly too low (to avoid overheating)”.2 Indeed, monitoring of blood gases, glucose, electrolytes, Silverman’s trial of adrenocorticotrophic hormone to infection indicators, and bilirubin; apnoea alarms; oxygen treat retinopathy of prematurity showed no benefi t and and ventilatory support; and antibiotics and other an increased death rate in the intervention group. He essential drugs for infection and shock. Similar reductions warned that we need trial evidence for any new preterm in case-fatality rate are highly unlikely in settings where intervention. No shortcuts can be used—care of the level 2 care is not available. Coverage of good-quality preterm infant is a delicate and integrated process. level 2 care in low-income settings is tiny—few facilities Recently, antenatal corticosteroid treatment has exist outside capital cities. One study of 1000 low- been widely promoted to reduce preterm deaths in birthweight infants in Dhaka, Bangladesh, showed that developing countries. On the face of it, the argument even in a teaching hospital that provided level 2 care, seems to be persuasive. Worldwide, 15 million infants three-quarters of neonates born at less than 33 weeks’ are born prematurely every year, and south Asia accounts gestation died during the neonatal period.6 The causes for two-thirds and Africa for three-quarters of deaths in of death were manifold and not just attributable to these infants.3 The 2013 State of the World’s Mothers respiratory distress. Indeed, respiratory distress syndrome report says: “Prenatal corticosteroids cost as little as might be less prevalent in poor countries than in wealthy 51 cents per treatment… are ready for rapid scale-up nations because intrauterine growth retardation now…using skilled birth attendants…and could save causes fetal cortisol concentrations to rise, which might 340 000 newborn lives each year”.4 This estimate does accelerate the production of lung surfactant. not come from trials but rather from a Lives Saved The second crucial question asks whether safety issues Analysis tool.4 exist in poor settings. If steroids are given to millions of The evidence is far more nuanced. A systematic women in preterm labour in poor populations, can we review of 21 studies in high-income and middle- be sure that there will be no signifi cantly increased risk of income country hospitals, which included a total of maternal sepsis, perinatal death, or childhood disability? 3885 women and 4269 infants, showed that one course The 2006 systematic review showed that, from eight of corticosteroids to accelerate fetal lung maturation trials and 1003 women with data for puerperal sepsis, in women at risk of preterm birth reduced neonatal 57 of 496 treated mothers had sepsis, compared deaths by 31% (relative risk [RR] 0·69, 95% CI 0·58–0·81; with 44 of 507 controls (RR 1·35, 95% CI 0·93–1·95). in 18 studies of 3956 infants) and respiratory distress More worrying is the evidence from trials that used syndrome by 34% (0·66, 95% CI 0·59–0·73; in 21 studies dexamethasone (the low-cost treatment recommended of 4038 infants).5 Nonetheless, the statement made for scale-up), which signifi cantly increased both by the Healthy Newborn Network that “administering puerperal sepsis (RR 1·74, 95% CI 1·04–2·89; in four For the Healthy Newborn antenatal steroids to a mom in preterm labour, helps trials of 536 women) and fever that needed antibiotics Network see www. healthynewbornnetwork.org/ her baby speed up lung development, reducing the risk (2·05, 1·14–3·69; in one trial of 118 women), in topic/antenatal-corticosteroids of newborn death by more than 50% in low-resource comparatively wealthy populations.7 These fi ndings
www.thelancet.com/lancetgh Published online March 13, 2014 http://dx.doi.org/10.1016/S2214-109X(14)70020-8 1 Comment
ring alarm bells for scale-up in populations in which soon be given steroid injections each year which carry the prevalence of malnutrition and the risk of sepsis are risks that could substantially exceed the benefi ts. much higher and access to antibiotics is low. In infants We support the existing recommendation to restrict born at a gestation of at least 36 weeks, an almost single-dose antenatal steroids to mothers who are at signifi cant trend was recorded towards an increase in 33 weeks’ gestation or less, in preterm labour, and with combined fetal and neonatal death (RR 3·25, 95% CI easy access to good quality, round-the-clock level 2 care. 0·99–10·66; in two studies of 498 infants). We need clear evidence for benefi t in other settings. Another major concern is that a substantially One multicountry trial at established research sites increased risk of disability can occur in cases where in Argentina, Guatemala, Kenya, India, Pakistan, and preterm survival is accompanied by suboptimum Zambia is underway.8 Other trials in poor and inaccessible care. Historically, outcomes for preterm infants born populations are needed before we can be certain of the at weights of less than 1500 g in developed countries risk– benefi t ratio. showed that mortality rates and the prevalence of major No quick fi x exists to ensure preterm survival without disability in survivors were very high until after 1960 serious disability. The global priority remains greater when level 2 care improved on a wide scale. coverage of round-the-clock provision of high-quality The third crucial question is whether there is any level 2 care. The scope for low-cost innovation in district evidence of benefi t for the great majority of preterm facilities is tremendous—eg, kangaroo care, the use of infants born at more than 33 weeks? A recent nursing aides to support scarce nursing staff , simple working paper for the UN Commission on Life-Saving technology for blood sugar and bilirubin estimation, Commodities for Women and Children promotes scale- electricity-free syringe pumps, and low-cost incubators up of steroids by suggesting “the eff ect is greatest for babies too ill to manage skin-to-skin care. However, between 31 weeks and 36 weeks gestation” although technology and drugs alone are dangerous without the no evidence shows an eff ect beyond 33 weeks.7 Three- provision and retention of skilled and motivated staff — quarters of the 12·6 million preterm infants worldwide the biggest challenge of all. each year are born after 33 completed weeks of gestation. In this group, the systematic review of antenatal Kishwar Azad, Anthony Costello* steroids5 showed no reduction in fetal or newborn Perinatal Care Project, Diabetic Association of Bangladesh, Dhaka, deaths. For deaths in newborn babies alone, there was Bangladesh (KA); Institute for Global Health, University College London, 30 Guilford Street, London WC1N 1EH, UK (AC) no benefi t for babies born at 34 weeks or more (RR 1·58, [email protected] 95% CI 0·71–3·50; two studies, 808 infants) or for babies We declare that we have no competing interests. born at 36 weeks or more (2·62, 95% CI 0·77–8·96, Copyright © Azad et al. Open Access article distributed under the terms of CC BY. three studies, 514 infants). Moreover, there was no 1 Silverman WA. Retrolental fi broplasia: a modern parable. New York: Grune reduction in respiratory distress at more than 34 weeks, & Stratton, 1980. 2 Silverman WA. Medicine’s ‘therapeutic imperative’ – haunted by the as we might expect from the timing of maturation of potential for tragedy. Paediatr Perinat Epidemiol 2004; 18: 318–19. surfactant production. For these reasons, the American 3 Lawn JE, Davidge R, Paul VK, et al. Born too soon: care for the preterm baby. Reprod Health 2013; 10 (suppl 1): S5. Academy of Obstetrics and Gynecology and the British 4 Save the Children. Surviving the fi rst day. State of the world’s mothers. Save Royal College of Obstetrics and Gynaecology both the Children, 2013. 5 Roberts D, Dalziel SR. Antenatal corticosteroids for accelerating fetal lung recommend steroid treatment for mothers only up to maturation for women at risk of preterm birth. Cochrane Database Syst Rev 33 weeks’ gestation. 2006; 3: CD004454. 6 Yasmin S, Paul E, Osrin D, Costello A. Neonatal mortality of low-birth- With no evidence of benefi t after 33 weeks in any weight infants in Bangladesh. Bull World Health Organ 2001; 79: 608–14. 7 Antenatal corticosteroids for the reduction of deaths in preterm babies. setting, questionable benefi t at less than 33 weeks in Working paper for the United Nations Commission on Life Saving poor populations, potential risks of sepsis in mothers, Commodities. March 2012. http://www.everywomaneverychild.org/images/ Final_AN_Steroids_report_for_UN-COMPLETE.pdf (accessed March 1, 2014). and unknown disability levels in babies in cases where 8 Althabe F, Belizán JM, Mazzoni A, et al. Antenatal corticosteroids trial in level 2 care standards are not met, we urge extreme preterm births to increase neonatal survival in developing countries: study protocol. Reprod Health 2012; 9: 22. caution. Millions of mothers in poor populations could
2 www.thelancet.com/lancetgh Published online March 13, 2014 http://dx.doi.org/10.1016/S2214-109X(14)70020-8