Background Essential and non-essential elements have an important impact on the development of the central nervous system during fetal development. Due to their less developed brain, preterm infants are more sensitive to element exposure, and are high-risk groups of neurodevelopmental abnormalities. However, it is not clear whether the effects of element exposure in utero on postpartum neurodevelopment are different between full-term infants and preterm infants. Objective To evaluate the effects of element exposure levels during pregnancy on neurodevelopment of children aged 6-24 months (of corrected age), and compare the effects between preterm and full-term children. Methods A prospective study design was adopted and this study was conducted based on the Maoming Birth Cohort Study (MBCS) in Maoming City, Guangdong Province. Twenty elements in cord blood of 197 preterm infants and 297 full-term infants were measured, including 11 essential trace elements [vanadium (V), chromium (Cr), manganese (Mn), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), selenium (Se), strontium (Sr), tin (Sn), and iron (Fe)], and 9 non-essential trace elements [aluminum (Al), arsenic (As), thallium (Tl), lead (Pb), uranium (U), cerium (Ce), antimony (Sb), cadmium (Cd), and yttrium (Y)]. The neurodevelopment of the children at 6, 12, and 24 months were evaluated by the Ages and Stages Questionnaires-the Third Edition (ASQ-3). A generalized estimating equation (GEE) model was adopted to evaluate the associations between elements and neurodevelopment in full-term and preterm children separately. Results The positive rates of 10 elements (Mn, Cu, Zn, Se, Sr, Fe, Sb, Tl, Pb, and As) in cord blood were greater than 80%. Among the preterm birth children, the results of GEE analysis showed that after adjusting for the covariates, for each increase of interquartile range (IQR) in ln-transformed concentration, As was associated with problems/delay in the communication and problem-solving sub-scales, with the adjusted odds ratios (OR) and 95% confidence intervals (CI) of 1.36 (1.03-1.80) and 1.55 (1.10-2.20), respectively; the adjusted OR (95%CI) of problems/delay in the fine motor and problem-solving sub-scales were 1.44 (1.00-2.07) and 1.76 (1.09-2.84) for Sb, respectively; the adjusted OR (95%CI) of problems/delay in the communication sub-scale was 1.37 (1.09-1.74) for Se. No statistically significant associations between umbilical cord blood element concentrations and neurodevelopment indicators were observed among full-term children. The results of stratified analysis by sex showed that the associations between umbilical cord blood element concentrations and neurodevelopment problems/delay were only significant among female preterm children. Conclusion Exposures to As, Se, and Sb during pregnancy may increase the risk of neurodevelopment problems/delay in preterm children aged 6-24 months, and female seem to be more vulnerable.