Beyond cut-offs: gestational age-specific perinatal mortality across the birthweight-for-gestational-age continuum-a population-based cross-sectional study.

Abnormal fetal growth is linked to increased perinatal morbidity and mortality. Birthweight percentiles-commonly the 10th and 90th-define small (SGA) and large for gestational age (LGA) infants. Although birthweight charts are periodically updated, cut-offs for SGA and LGA remain fixed, despite changes in population characteristics and evolving clinical management. We aimed to characterize how perinatal mortality varies continuously across birthweight-for-gestational-age (BW-for-GA) and gestational age (GA), describing GA-specific mortality gradients rather than defining new cut-offs. We analyzed 2,581,816 singleton infants born in the Netherlands (2000-2015), excluding congenital malformations, extreme birthweights, or missing data. Perinatal mortality was defined as fetal or early neonatal death (< 7 days). Logistic regression modeled mortality based on continuous GA and BW-for-GA percentile, with rate ratios calculated relative to the 50th percentile and GA-specific nadir. Sensitivity analyses excluded antepartum stillbirths. Perinatal mortality declined gradually with GA, reaching a nadir at 40 weeks. Across BW-for-GA, mortality was lowest at the 80th-83rd percentile and higher at both extremes. Infants below the 3rd percentile had at least a twofold increase in mortality compared with those at median birthweight, while comparable increases for infants below the 10th percentile were observed only after 37 weeks. Sensitivity analyses excluding antepartum stillbirths reduced overall mortality but preserved the curvilinear pattern.

conclusionPerinatal mortality varies continuously, without discrete inflection points, revealing graded, gestational-age-specific patterns across the population and highlighting the limitations of rigid SGA and LGA thresholds. WHAT IS KNOWN: • SGA and LGA infants face higher risks of adverse outcomes. • Although birthweight charts are regularly updated, fixed percentile cut-offs such as SGA and LGA are rarely reconsidered despite changing populations and evolving clinical management. WHAT IS NEW: • Perinatal mortality varies continuously across birthweight-for-gestational-age, without discrete inflection points, revealing graded, GA-specific patterns. • This highlights the limitations of fixed SGA/LGA percentile thresholds and supports nuanced interpretation of risk for clinical decision-making.