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 Randomized Trial

Abstract

Antenatal corticosteroids are 1 of the most effective hospital-based treatments to reduce neonatal mortality for pregnant women at risk of preterm birth. While studies have shown reductions in neonatal mortality in high-and middle-income countries, and the treatment has been recommended by national and international health organizations, less than 10% of women in low-income countries at risk of preterm birth receive antenatal corticosteroids. This cluster-randomized trial aimed to assess the feasibility, safety, and effectiveness of a multifaceted intervention to increase the use of these treatments in middle-and low-income countries. Between October 2011 and March 2014, this trial randomly assigned (1: 1) clusters in rural and semiurban areas in Guatemala, Argentina, India, Kenya, Pakistan, and Zambia to this intervention or to standard care. To be eligible, clusters had to have at least 300 births annually in the defined area, either at homes or in facilities. Part of the intervention included better identification of women at risk of preterm birth. In intervention clusters, all health care providers were trained to identify women at risk of preterm birth and to administer dexamethasone in 1 course of four 6-mg doses every 12 hours to at risk women at less than 36 weeks' gestational age. In the intervention arm, women identified as high risk of preterm birth were also referred, but transportation was not provided. All clusters were training in newborn care. There were 51 intervention clusters with 2520 live births less than the fifth percentile for birth weight and 50 control clusters with 2258 in that percentile. In the intervention cluster, 45% of women received antenatal corticosteroids, whereas 10% of the control group received them (P < 0.0001). Of the less-than-fifth-percentile infants in the intervention group, 28-day neonatal mortality was 225 per 1000 live births, compared with a 323-per-1000-live-births neonatal mortality rate in that percentile in the control group (relative risk (RR), 0.96; 95% confidence interval [CI], 0.87-1.06; P = 0.65). Stillbirths occurred in 229 per 1000 births in the intervention group and 247 per 1000 births in the control group (RR, 0.99; 95% CI, 0.90-1.09; P = 0.81). Including all pregnancies, the neonatal mortality rate for the intervention group was 27.4 per 1000 and 23.9 per 1000 for the control group (RR, 1.12; 95% CI, 1.02-1.22; P = 0.0127), and the stillbirth rate in the intervention group was 26.8 per 1000, and that rate was 24.3 per 1000 in the control group (RR, 1.11; 95% CI, 1.02-1.22; P = 0.0181). Researchers found that the intervention group was overtreated with antenatal corticosteroids, and even though nearly half of women delivering less-than-fifth-percentile infants received steroids, neonatal mortality did not decrease. Neonatal deaths were significantly increased in the intervention group (3.5 per 1000 live births), as were perinatal deaths (5.1 per 1000 live births) and maternal infection (3%-6% increase). The authors of the study advise caution in implementing interventions such as this one. The findings do not support this strategy, and whether the antenatal corticosteroids directly caused the unfavorable results in the intervention group should be further explored.