BackgroundPrevious studies have shown maternal hepatitis B virus infection is associated with preterm birth. However, whether liver dysfunction caused by hepatitis B virus infection or other factors can lead to preterm delivery needs further exploration. Serum alanine aminotransferase (ALT) is a sensitive indicator of liver function, but few studies have investigated the association between serum ALT concentrations and preterm birth. We aimed to examine the relationship between maternal preconception ALT concentration and risk of preterm birth. MethodsWe did a population-based cohort study of 4832887 women aged 20–49 years old, who participated in the National Free Pre-pregnancy Checkups Project (NFPCP) in 2013–16, and delivered a livebirth before Dec 31, 2017, in rural China. We collected information about demographic characteristics, history of pregnancy and diseases, serum ALT concentration, hepatitis B surface antigen (HBsAg) status, and other variables. We divided participants into five groups according to serum ALT concentration (≤20 U/L, 21–40 U/L, 41–60 U/L, 61–80 U/L, and ≥ 81 U/L). The primary outcome was preterm birth (gestational age 28 to <37 weeks). We used logistic models to estimate odds ratios (ORs) and 95% CIs, after adjusting for confounding variables. We applied restricted cubic splines to investigate the dose-response relationship. The study was approved by the Institutional Research Review Board at the National Health Commission, and all participants gave written consent. FindingsBy Dec 31, 2017, 324207 (6·71%) preterm birth events were documented in the 4832887 participants. Incidence of preterm birth was lowest in women with ALT concentrations of 21–40 U/L (93685 [6·42%] of 1 458 233 women), whereas the preterm birth incidence was highest in women with ALT concentrations of 81 U/L or higher (3302 [7·52%] of 43912 women). There was a non-linear dose-response relationship between serum ALT concentration and risk of preterm birth (U-shaped, p<0·001). In women who were HBsAg negative, the multivariable-adjusted OR for preterm birth was 1·07 (95% CI 1·06–1·08) for ALT concentrations of 20 U/L or less, 1·19 (1·16–1·22) for concentrations of 41–60 U/L, 1·16 (1·11–1·20) for concentrations of 61–80 U/L, and 1·16 (1·12–1·21) for concentrations of 81 U/L or higher, when compared to women with ALT concentrations of 21–40 U/L. In women who were HBsAg positive, there were significant associations between preterm birth and ALT concentrations of 61–80 U/L (OR 1·17, 95% CI 1·05–1·29), and 81 U/L or higher (1·21, 1·11–1·31). InterpretationOur study identified a U-shaped relationship between maternal preconception ALT concentration and risk of preterm birth. The association between ALT concentration and risk of preterm birth was independent of hepatitis B virus infection. The surveillance of ALT among women planning pregnancy should be warranted, given the increased risk of preterm birth. FundingThis study was supported by the National Key Research and Development Program (number 2016YFC10003007).