Antenatal Dexamethasone Reduces Mortality in Preterm Infants in Low Resource Countries

January 14, 2022

MANUSCRIPT CITATION

Oladapo, OT, et al. (2020). “Antenatal Dexamethasone for Early Preterm Birth in Low-Resource Countries” The New England Journal of Medicine. PMID: 33095526

REVIEWED BY

Monika S. Patil, MD, FAAP
Assistant Professor of Pediatrics
Baylor College of Medicine
monika.patil@bcm.edu

Bheru B. Gandhi, MD, FAAP
Assistant Professor of Pediatrics
Baylor College of Medicine
Bheru.gandhi@bcm.edu

Sharada H. Gowda, MD FAAP
Assistant Professor of Pediatrics
Baylor College of Medicine
shgowda@bcm.edu

TYPE OF INVESTIGATION

Prevention

QUESTION

Do (P) premature infants born to mothers at risk for preterm delivery who are given (I) antenatal dexamethasone compared to (C) placebo have (O) reduced neonatal mortality and combined stillbirth and neonatal mortality, without resulting an increased maternal infection risk?

METHODS

  • Design: Prospective multinational, multicentre, parallel-group, double-blind, individually randomized, placebo-controlled trial
  • Allocation: Computer generated randomization with site stratified balanced blocks of 10. Enrolled participants at each site were then randomized 1:1.
  • Blinding: Patients, care providers, and investigators were all blinded to trial-group assignments
  • Follow-up period: follow up of the women was conducted until 28 days after they gave birth or until death (whichever occurred first), and follow up of the fetuses occurred until 28 days after birth or until death (whichever occurred first).
  • Setting: 29 secondary and teritiary level hospitals across six trial sites in five countries (Bangladesh, India, Kenya, Nigeria -2 sites, and Pakistan).
  • Patients:
    • Inclusion:
      • Pregnant women with confirmed live fetuses between 26 weeks 0 days and 33 weeks 6 days of gestation, and at risk for preterm birth planned or expected in the next 48 hours.
      • Written informed consent obtained from the pregnant woman before enrollment
    • Exclusion criteria:
      • Clinical signs of severe infection
      • Major congenital fetal anomalies
      • Concurrent or recent use of systemic glucocorticoids
      • Contraindication to receiving glucocorticoids
      • Participating in another trial
  • Intervention:
    • Dexamethasone group: Dexamethasone intramuscular injections 6mg every 6 hours for 4 doses
    • Control Group: Identical appearing placebo intramuscular injection every 6 hours for 4 doses
    • Women were eligible for a second course of Dexamethasone or Placebo (course matching their initial assignment) if they had not given birth after 7 days, and still met the inclusion criteria
  • Outcomes:
    • Primary outcomes:
      • Neonatal death (death of a live-born infant within 28 completed days of life)
      • Stillbirth (death of a fetus post randomization) or neonatal death
      • Composite of possible maternal bacterial infection (maternal fever- temperature >38 degrees C, or clinically suspected or confirmed infection for which therapeutic antibiotics were used
    • Secondary outcomes:
      • Neonatal:
        • Stillbirth
        • Early neonatal death (Death of a live birth within 7 completed days of life)
        • Severe respiratory distress (Clinical features such as the presence of fast breathing respiratory rate ≥ 70 breaths per minute, SpO2 less than 90%, or use of supplemental oxygen, AND at least one of the following clinical signs: 1. Marked nasal flaring during inspiration, 2. Expiratory grunting audible with naked ear 3. Severe chest in drawing)
        • Neonatal sepsis (Defined as the presence of at least two (or more) of the following signs: Stopped feeding well, Severe chest in-drawing, Fever (body temperature of 38 °C or greater), Hypothermia (body temperature less than 35.5 °C), Movement only when stimulated or no movement at all, Convulsions)
          • Severe Intraventricular hemorrhage (sIVH) (Defined as a Papile’s intraventricular hemorrhage classification grade 3 or 4, as per transcranial ultrasound assessment)
          • Neonatal hypoglycaemia (blood glucose measure less than 45 mg/dl (2.6mmol/l))
          • Apgar score at 5 minutes
      • Maternal:
        • Maternal death (Any maternal death in a trial participant, from time of randomization to 28 completed days postpartum)
        • Maternal fever (fever ≥38.0 C since randomization during hospital admission/s only)
        • Chorioamnionitis (suspected or confirmed based on clinical assessment by obstetric care physician during hospital admission/s only)
        • Postpartum endometritis (suspected or confirmed based on clinical assessment by obstetric care physician during hospital admission/s only)
        • Wound infection (Infection of a wound or incision site suspected or confirmed by obstetric care physician during hospital admission/s only)
        • Non-obstetric infection (suspected or confirmed based on clinical assessment by obstetric care physician during hospital admission/s only: Respiratory tract infection (including pneumonia, pharyngitis, sinusitis or similar), Urinary tract infection, Pyelonephritis, Acute cholecystitis, Other systemic infection. Malaria was specifically excluded from this outcome)
  • Analysis and Sample Size:
    • Sample size was determined on the basis of neonatal death as the primary outcome targeting a decrease of 15% or more in a two sided 5% significance test with 90% power and accounting for 10% loss to follow up. A recruitment of 6018 women was estimated to be the sample size.
    • Data were analyzed for the neonatal primary outcomes with the intention to treat analysis as recommended for a superiority hypothesis. Maternal infection was analyzed with per-protocol analysis as recommended for a non-inferiority hypothesis.
    • A sample size of 5416 women was estimated targeting a decrease in the risk if neonatal death of > 15%, from 25% to 21.25%. With an estimated 10% patient loss to follow up, the target sample size was 6018 (α = 0.05, β = 0.80).
    • Intention to treat analysis was performed for all primary outcomes. A secondary per-protocol analysis was peformed for the maternal primary outcome which excluded women with protocol violations.
    • For the primary outcomes of fetal/neonatal mortality and maternal severe infection, analysis was done on the enrolled population but for neonatal mortality, analysis was done only on liveborn neonates.
    • Data analysis was first performed on all available data and then to account for missing data, sensitivity analysis was done through multiple imuptation. Analysis with the false-discovery-rate approach was also to correct for multiplictity further strengthening the significant findings.
    • Primary outcome data from the dexamethasone group was compared to the placebo group using logistic regression to determine the relative risk. Results were also stratified by trial site. Primary outcomes were analyzed in pre-specified subgroups.
    • Three interim analysis were planned by the data and safety monitoring board but the trial was stopped after the second interim analysis of 2304 women and 2536 fetuses and infants with complete follow up of primary outcomes.
  • Patient Follow up:
    • 7008 women were screened for eligibility, and 2852 underwent randomization. 4156 women were excluded because they did not meet eligibility criteria by the time of birth, or did not give consent. 1429 women (with 1544 fetuses) were randomized into the dexamethasone group, and 1423 women (with 1526 fetuses) were randomized into the placebo group. Some women were lost to follow up (n=17), declined to participate or withdrew after randomization (n=4), or had infants that could not be located for follow up (n=15). A total of 1416 women, and 1532 fetuses and infants were assessed for the primary outcomes in the dexamethasone group. A total of 1412 women, and 1519 fetuses and infants were assessed for primary outcomes in the control group.

MAIN RESULTS

All the women received at least one dose of dexamethasone or placebo according to their randomization. In the dexamethasone group 57% of women received 4 doses, and in the placebo group 53% of women received 4 doses. Baseline characteristics of both groups were similar at trial entry.

Primary outcome: The incidence of neonatal death was significantly lower in the dexamethasone group compared to the placebo group (19.6% vs 23.5%; RR 0.84; 95% confidence interval [CI], 0.72 to 0.97; P=0.03). The incidence of stillbirth or neonatal death was also significantly lower in the dexamethasone group compared to the placebo group (25.7% vs 29.2; RR 0.88; 95% CI 0.78 to 0.99, P=0.04). Possible maternal bacterial infection was not higher in the dexamethasone group compared to the placebo group (4.8% vs 6.3%; RR 0.76; 95% CI 0.56 to 1.03; P=0.002 for non-inferiority).

Secondary outcome: The incidence of early neonatal death, severe respiratory distress syndrome 24 hours after birth, neonatal hypoglycemia 6 hours after birth, the use of CPAP, and the duration of oxygen therapy were all significantly lower in the dexamethasone group compared to the placebo group (table 1). Other secondary neonatal outcomes, and secondary maternal outcomes were similar in both groups.

CONCLUSION

Authors conclude that using antenatal dexamethasone for women at risk for imminent preterm birth in low resource settings significantly reduces the risks of neonatal death, and stillbirth or neonatal death, without increased harm to women or newborns. Authors advocate for dexamethasone use in hospital settings where obstetric physicians can confirm the gestational age and identify women at risk for imminent preterm birth, and where resources for providing minimum standard neonatal care for premature newborns are available, including access to CPAP and oxygen.

COMMENTARY

Antenatal corticosteroids (ANC) improve survival and decrease morbidity in premature infants [1, 2]. However, their use has been controversial in low resource countries (LRC) due to the ACT study which questioned the safety and efficacy of ANC [3]. The ACT study had limitations; it did not collect or process data on the use of co-interventions affecting outcomes thus confounding the results. They used birthweight percentiles instead of gestational age to define the target subgroup thus misclassifying some term infants as preterm. To answer the question better, Oladapo et al., the WHO ACTION collaborators, designed a multi-country randomized control blinded study. This study found significant reduction in neonatal death and stillbirth and death with the use of antenatal dexamethasone without increasing maternal morbidity or mortality. The researchers adhered to stricter inclusion and exclusion criteria. They obtained accurate gestational age ultrasound dating, enrolled women who were at risk for preterm birth within 48 hours, and delivered in hospitals equipped with essential standard resources to support preterm newborns.

Previous studies show benefit for premature newborns exposed to ANC in LRC [4, 5], but these studies are not placebo controlled. Strengths of this study include the size, randomization, blinding, and multi-national. The study was designed to identify and recruit patients and fetuses who would potentially receive the most benefit from ANC. Reporting data with and without use of multiple imputation and using the false-discovery-rate approach amplified the significance of findings. Additionally, because the primary outcome in this study was neonatal death, not respiratory distress syndrome or other morbidities, the crucial impact of ANC on reducing preterm mortality in LRC is apparent.

One weakness is the clinical management of the mother-baby dyad varied depending on local resources and guidelines. This may alter outcomes depending on the training and protocols of a particular center. Data were analyzed with and without stratifying for study site. Despite this, there was still benefit to preterm newborns exposed to antenatal dexamethasone demonstrating its favorable impact on mortality and morbidity which is important as it will be used in many LRC.

Prematurity is the leading cause of death in children under age 5 worldwide [6]. Low-cost, high impact preventive interventions are needed to help reduce this burden. This study reaffirms that dexamethasone is safe, efficacious, and a cost saving for  LRC to reduce their hospital based stillbirth or neonatal mortality in premature infants. Dexamethasone has the potential to make a major impact for the survival of premature newborns in LRC when used with the appropriate clinical protocolized approach.

REFERENCES

  1. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of the respiratory distress syndrome in premature infants. Pediatrics 1972; 50(4): 515-25.
  2. McGoldrick E, Stewart F, Parker R, Dalziel SR. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. Cochrane Database Syst Rev 2020; 12: CD004454.
  3. Althabe F, Belizan JM, McClure EM, Hemingway-Foday J, Berrueta M, Mazzoni A et al. A population-based, multifaceted strategy to implement antenatal corticosteroid treatment versus standard care for the reduction of neonatal mortality due to preterm birth in low-income and middle-income countries: the ACT cluster-randomised trial. Lancet 2015; 385(9968): 629-639.
  4. Fekih, M, Chaieb A, Sboui H, Denguezli W, Hidar S, Khairi H. Value of prenatal corticotherapy in the prevention of hyaline membrane disease in premature infants. Randomized prospective study. Tunis Med 2002; 80(5): 260-5.
  5. Qublan HS, Malkawi HY, Hisat MS, Hindawi IM, Al-Taani MI, Abu-Khait SA et al. The effect of antenatal corticosteroid therapy on pregnancies complicated by premature rupture of membranes. Clin Exp Obstet Gynecol 2001; 28(3): 183-6.
  6. WHO. WHO Global Health Observatory Data. 2021 [cited 2021; Available from: https://www.who.int/data/gho/data/themes/topics/indicator-groups/indicator-group-details/GHO/causes-of-child-death.
css.php