Maternal hemoglobin concentration measured in early pregnancy is associated with the risk of severe neonatal morbidity and mortality, according to a large population-based study published in Annals of Internal Medicine. The findings indicate that both low and high hemoglobin levels in the first trimester are linked to adverse neonatal outcomes.
The retrospective cohort study included 1,100,341 singleton births in Ontario, Canada, between 2007 and 2023. All included pregnancies had a hemoglobin measurement performed between 2 and 12 weeks’ gestation, before substantial pregnancy-related plasma volume expansion. Researchers examined the association between early pregnancy hemoglobin concentration and severe neonatal morbidity and mortality (SNM-M), a validated composite outcome capturing major neonatal diagnoses, critical interventions, and death within 27 days of birth.
Hemoglobin concentration was analyzed as a continuous variable, with 125 g/L used as the reference value. Models were adjusted for maternal age, parity, socioeconomic indicators, tobacco or substance use, and preexisting conditions including hypertension, diabetes mellitus, chronic kidney disease, systemic lupus erythematosus, and inflammatory bowel disease.
The analysis showed a U-shaped relationship between hemoglobin concentration and SNM-M. Compared with the reference value of 125 g/L, lower hemoglobin levels were associated with higher risk. The adjusted relative risk of SNM-M was 1.08 at 105 g/L and 1.17 at 90 g/L. Elevated hemoglobin levels were also associated with increased risk, with adjusted relative risks of 1.05 at 135 g/L and 1.20 at 150 g/L.
At the reference hemoglobin level, the rate of SNM-M was 6.7%. Severe neonatal morbidity included outcomes such as extreme preterm birth, very low birthweight, respiratory distress syndrome, seizures, intraventricular hemorrhage, hypoxic-ischemic encephalopathy, sepsis, ventilatory support, and newborn resuscitation.
Secondary analyses demonstrated similar patterns for preterm birth and for combined outcomes of SNM-M or stillbirth. The association persisted after excluding extreme preterm birth and very low birthweight from the composite outcome. Further adjustment for maternal body mass index and hyperemesis gravidarum did not materially change the results.
In a subset of pregnancies with available ferritin measurements, joint analysis of hemoglobin and iron status suggested a modest increase in risk at low ferritin levels, though iron status could not be assessed comprehensively across the cohort.
The authors note that measuring hemoglobin early in pregnancy minimizes confounding from later physiologic hemodilution. Limitations include reliance on administrative data and lack of information on iron supplementation.
Disclosure forms are available with the article online.
Source: Annals of Internal Medicine
