Perfluoroalkyl substances (PFAS) are persistent bio-accumulative chemicals that impact the health of pregnant women and their children. PFAS derive from environmental and consumer products, which depend on human lifestyle, socioeconomic characteristics, and time variation. Here, we aimed to explore the temporal trends of PFAS in pregnant women and the characteristics related to maternal PFAS concentration. Our study is part of the Hokkaido Study on Environment and Children’s Health, the Hokkaido large-scale cohort that recruited pregnant women between 2003 and 2011. Blood samples were acquired from pregnant women during the third trimester to measure PFAS and cotinine concentrations. Maternal basic information was collected with a baseline structured questionnaire. Eleven PFAS were measured from 2123 samples with ultra-performance liquid chromatography coupled with triple quadrupole tandem mass spectrometry. Eight PFAS were above 80% detection rate and were included in the final analysis. We used multivariable linear regression to analyze the association of pregnant women characteristics with the levels of eight PFAS. The temporal trend of PFAS was observed in two periods (August 2003 to January 2006 and February 2006 to July 2012). The concentration of perfluorooctane sulfonate (PFOS) significantly decreased from August 2003 to January 2006 and from February 2006 to July 2012. The concentrations of perfluorododecanoic acid (PFDoDA), perfluoroundecanoic acid (PFUnDA), and perfluorotridecanoic acid (PFTrDA) increased significantly between August 2003 and January 2006, whereas they decreased significantly between February 2006 and July 2012. Women with pre-pregnancy body mass index (BMI) >25 kg/m2 had lower PFUnDA, PFDoDA, and PFTrDA levels than did those with normal BMI (18.5–24.9 kg/m2). Pregnant women, who were active smokers (cotinine > 11.49 ng/mL), had higher PFOS than the non-smokers (cotinine < 0.22 ng/mL). Lower levels of PFHxS, PFOS, PFOA, PFNA, and PFDA were observed in women, who had given birth to more than one child. There were also significant positive associations between PFAS levels and annual income or maternal education. PFAS levels varied in women with higher pre-pregnancy BMI, active smoking status, higher education level and annual income. The causes of the individual PFAS differences should be explored in an independent study.