Background and aimPerfluoroalkyl and polyfluoroalkyl substances (PFAS) have demonstrated potential toxicity in skeletal development. However, the relationship between prenatal PFAS exposure and offspring bone health remains unclear in epidemiological studies. Therefore, we aim to investigate whether prenatal exposure to PFAS is associated with bone mineral density (BMD) in offspring. MethodThis study population included 182 mother-child pairs in the Shanghai Obesity and Allergy Cohort, enrolled during 2012–2013. 10 PFAS were measured by liquid chromatography-mass spectrometry (LC-MS) in cord plasma. The child's spinal BMD was measured using a dual-energy X-ray absorptiometry (DXA) scanner at the age of 8. Multivariable linear regression models were used to estimate the associations between individual PFAS concentrations (as a continuous variable or categorized into quartiles) and child BMD. Bayesian kernel machine regression (BKMR) was employed to explore the joint effects of PFAS mixtures on BMD. ResultsAmong the 10 PFAS, 8 of them had a detection rate >90% and were included in the subsequent analysis. We observed no significant associations between individual PFAS (as a continuous variable) and spinal BMD in 8-year-old children using the multivariable linear regression model. When treated as quartile categories, the second and fourth quartiles of perfluorobutane sulfonate (PFBS) was associated with higher BMD in the first lumbar vertebra, compared with the lowest quartile. BKMR analysis revealed no association between the PFAS mixture and child BMD. ConclusionWe observed no associations of prenatal PFAS exposure with child BMD at 8 years of age. Given the inconsistent epidemiological evidence, further research is needed to confirm these findings from other studies or elucidate the potentially toxic effects of PFAS on bone.