Whether plasma serotonin (5-HT) levels could be a biomarker for postpartum depression (PPD) diagnosis is under dispute.A total of 979 of pregnant women without antenatal depression at the time of delivery (TD) were enrolled and followed up at six weeks postpartum (SWP) in Changsha, China. The odds ratio (OR) and 95% confidence interval (CI) for plasma 5-HT level at TD, at SWP, changes in 5-HT, and risk of PPD and deterioration in EPDS scores at SWP were estimated by Logistic regressions. Restricted cubic spline (RCS) functions were also used to assess the dose-response relationships.The 6-week cumulative incidence of PPD was 12.05% (95%CI:10.08%, 14.26%). The average level of plasma 5-HT changed from 223.65 ± 131.47 ng/ml at TD to 216.43 ± 122.73 ng/ml at SWP, with an average change of -7.22 ± 96.54 ng/ml. Plasma 5-HT at TD was negatively correlated with EPDS score at TD and SWP (p 0.05), as was the correlation between 5-HT at SWP and EPDS scores at SWP (p = 0.038). However, the changes in 5-HT were not associated with the EPDS score at SWP (p = 0.346). Neither plasma 5-HT level at TD nor changes in 5-HT was associated with PPD at SWP or deterioration in EPDS scores (p 0.05). Plasma 5-HT at delivery had insignificant discriminatory power for diagnosing PPD and prediction of deterioration in EPDS scores (p ≥ 0.05).Plasma 5-HT level at delivery was associated with EPDS score at delivery and SWP, but not with PPD at SWP suggesting that plasma 5-HT is not suitable as PPD diagnostic biomarker.