Stillbirth risk by fetal size among 126.5 million births in 15 countries from 2000 to 2020: A fetuses-at-risk approach

Abstract

Objective: To compare stillbirth rates and risks for small for gestational age (SGA), large for gestational age (LGA) and appropriate for gestational age (AGA) pregnancies at 24-44 completed weeks of gestation using a birth-based and fetuses-at-risk approachs.

Design: Population-based, multi-country study.

Setting: National data systems in 15 high- and middle-income countries.

Population: Live births and stillbirths.

Methods: A total of 151 country-years of data, including 126 543 070 births across 15 countries from 2000 to 2020, were compiled. Births were categorised into SGA, AGA and LGA using INTERGROWTH-21st standards. Gestation-specific stillbirth rates, with total births as the denominator, and gestation-specific stillbirth risks, with fetuses still in utero as the denominator, were calculated from 24 to 44 weeks of gestation.

Main outcome measures: Gestation-specific stillbirth rates and risks according to size at birth.

Results: The overall stillbirth rate was 4.22 per 1000 total births (95% CI 4.22-4.23) across all gestations. Applying the birth-based approach, the stillbirth rates were highest at 24 weeks of gestation, with 621.6 per 1000 total births (95% CI 620.9-622.2) for SGA pregnancies, 298.4 per 1000 total births (95% CI 298.1-298.7) for AGA pregnancies and 338.5 per 1000 total births (95% CI 337.9-339.0) for LGA pregnancies. Applying the fetuses-at-risk approach, the gestation-specific stillbirth risk was highest for SGA pregnancies (1.3-1.4 per 1000 fetuses at risk) prior to 29 weeks of gestation. The risk remained stable between 30 and 34 weeks of gestation, and then increased gradually from 35 weeks of gestation to the highest rate of 8.4 per 1000 fetuses at risk (95% CI 8.3-8.4) at ≥42 weeks of gestation. The stillbirth risk ratio (RR) was consistently high for SGA compared with AGA pregnancies, with the highest RR observed at ≥42 weeks of gestation (RR 9.2, 95% CI 15.2-13.2), and with the lowest RR observed at 24 weeks of gestation (RR 3.1, 95% CI 1.9-4.3). The stillbirth RR was also consistently high for SGA compared with AGA pregnancies across all countries, with national variability ranging from RR 0.70 (95% CI 0.43-0.97) in Mexico to RR 8.6 (95% CI 8.1-9.1) in Uruguay. No increased risk for LGA pregnancies was observed.

Conclusions: Small for gestational age (SGA) was strongly associated with stillbirth risk in this study based on high-quality data from high- and middle-income countries. The highest RRs were seen in preterm gestations, with two-thirds of the stillbirths born as preterm births. To advance our understanding of stillbirth, further analyses should be conducted using high-quality data sets from low-income settings, particularly those with relatively high rates of SGA.

Keywords: fetuses‐at‐risk approach; gestational age; preterm birth; size for gestational age; stillbirth.

FIGURE 1

(A) Flow chart of stillbirth data inclusion and exclusion. (B) Number of stillbirths included by country (n = 534 956) for 24–44 weeks of gestation.

FIGURE 2

(A) Distribution of size for gestational age (for 24–44 weeks of gestation) among all stillbirths. (B) Distribution of size for gestational age (for 24–44 weeks of gestation) among all live births.

FIGURE 3

(A) Stillbirth rate for small for gestational age (SGA), appropriate for gestational age (AGA) and large for gestational age (LGA) pregnancies by gestational week using a birth‐based approach across 15 countries, 2000–2020. (B) Stillbirth risk for SGA, AGA and LGA pregnancies by gestational week using a fetuses‐at‐risk approach across 15 countries, 2000–2020.

FIGURE 4

(A) Stillbirth risk ratio comparing small for gestational age (SGA) versus appropriate for gestational age (AGA) and comparing SGA versus large for gestational age (LGA) pregnancies by gestational week using a fetuses‐at‐risk approach across 15 countries, 2000–2020. (B) Stillbirth risk ratio comparing SGA versus AGA and comparing LGA versus AGA pregnancies by country using a fetuses‐at‐risk approach across all gestations (24–44 weeks of gestation), 2000–2020. The square symbol represents the risk ratio for each country, and the size of the square is proportional to the study weight. The whiskers extending from each side of the square represent the range of the 95% confidence interval (95% CI). The diamond symbol indicates the overall pooled effect size with a random‐effect model, which is centred at the point estimate, and the width of the diamond represents the 95% CI. Estonia and Lebanon were excluded from the meta‐analysis because of wide 95% CIs, as their inclusion would compromise the quality of the forest plot.