[Study of factors associated with the false-positive rate of second-trimester serological screening in 632,825 cases in Sichuan based on propensity score matching].

Objective: To retrospectively analyze the results of second-trimester serological prenatal screening and explore the factors which may influence the false-positive rate (FPR).

Methods: From January 2013 to December 2022, false-positive samples with follow-up outcomes from 632,825 second-trimester serological prenatal screening samples tested at Sichuan Provincial Maternity and Child Health Care Hospital were selected as the study group, while true-negative samples were 1 : 1 matched as the control group by propensity-score matching (PSM). Univariate and binary logistic regression models were used to analyze the influencing factors. This study was approved by the Medical Ethic Committee of Sichuan Provincial Maternity and Child Health Care Hospital (Ethic No. 20240607-270).

Results: The study and control groups were each matched with 305,998 cases. Univariate analysis showed that sampling season, the difference between ultrasound and gestational weeks calculated by last menstrual period (LMP), monthly median multiple of the median (mMoM) of alpha-fetoprotein (AFP), and monthly mMoM of free β -human chorionic gonadotropin (free β -hCG) were significantly different between the two groups (P < 0.05). Binary logistic regression analysis showed that Winter (OR = 0.938; 95%CI: 0.893 ~ 0.985), monthly AFP mMoM ≥ 1.11 (OR = 0.846; 95%CI: 0.761 ~ 0.941), monthly free β -hCG mMoM ≥ 0.89 (OR = 0.827; 95%CI: 0.737 ~ 0.929) are protective factors for FPR increase, whilst Spring (OR = 1.124; 95%CI: 1.072 ~ 1.179), Summer (OR = 1.121; 95%CI: 1.062 ~ 1.183), the difference between ultrasound and gestational weeks calculated by LMP of 8 ~ 14 days (OR = 1.319; 95%CI: 1.241 ~ 1.402), < 14 days (OR = 1.689; 95%CI: 1.542 ~ 1.850), monthly AFP mMoM of 0.90 ~ 0.94 (OR = 1.088; 95%CI: 1.046 ~ 1.131), and monthly free β -hCG mMoM of 1.05 ~ 1.10 (OR = 1.046; 95%CI: 1.000 ~ 1.094), ≥ 1.11 (OR = 1.062; 95%CI: 1.002 ~ 1.126) are risk factors for FPR increase.

Conclusion: Sampling season, difference between ultrasound and gestational weeks by LMP, monthly AFP mMoM, and monthly free β -hCG mMoM are risk factors for FPR during serological prenatal screening. Screening laboratories should look for the cause of abnormal FPR through such factors and adjust them accordingly.