Objective: To study the cognitive function status, serum phosphorylated tau181 (P-tau181) protein level, and total tau (T-tau) protein level in patients with preeclampsia (PE), pregnant healthy controls (PHCs), and non-pregnant healthy controls (NPHCs), and to research their feasibility as serum biomarkers for evaluating cognitive functional impairment in PE patients. Methods: Sixty-eight patients with PE, 48 NPHCs, and 30 PHCs were included. Cognitive functional status was assessed using standardized Symbol Digit Modalities Test (SDMT) and Montreal Cognitive Assessment (MoCA) scales. Enzyme-linked immunosorbent assay (ELISA) was used to detect the level of serum P-tau181 and T-tau protein. The concentration of serum P-tau181 and T-tau protein were compared by one-way analysis of variance in the three groups of subjects. The correlation between P-tau181, T-tau, and SDMT was explore by multiple linear regression analysis. The areas under the receiver operating characteristic (ROC) curves of serum P-tau181 and SDMT were calculated to predict the cognitive level of subjects. Results: PE patients significantly had lower scores on SDMT (47.97 ± 7.54) and MoCA (28.00 ± 2.00) than normotensive PHCs (30.00 ± 1.25, 54.73 ± 8.55, respectively). The significant difference was found in serum P-tau181 protein levels among the three groups [H(K) = 19.101, P < 0.001]. Serum P-tau181 was thicker in PE patients than PHCs or NPHCs (both P < 0.05). According to the ROC curve, T-tau had no statistical significance in predicting the ability of cognizance, while P-tau181 and SDMT had. The DeLong test showed that P-tau181 was better than T-tau in predicting the ability of cognizance (P < 0.05). Conclusion: The patients with PE have occurred the decline of cognitive function during pregnancy. The high level of serum P-tau181 can be used as a clinical laboratory indication for non-invasive assessment of cognitive functional impairment in PE patients.