Objective: Preventing adverse events due to unstable oxygen saturation (SpO2) at night in pregnant women is of utmost importance. Poor sleep has been demonstrated to impact SpO2 levels. Nowadays, many gravida have a habit of prolonged exposure to light before sleep, which can disrupt their sleep. Therefore, this study aimed at investigate the relationship between lights-out time, sleep parameters and SpO2, exploring the underlying mechanisms. Methods: The data of 2881 eligible subjects from the Nulliparous Pregnancy Outcomes Study Monitoring Mothers-to-be and Sleep Disordered Breathing (nuMOM2b-SDB) database were analyzed. Multiple linear regression models were used to investigate the relationship between lights-out time and SpO2. In addition, restricted cubic splines (RCS) were employed to fit the nonlinear correlation between the two variables. The smoothing curve method was further utilized to depict the relationship between lights-out time and SpO2 based on various subgroup variables. Results: All participants were categorized according to race/ethnicity. A negative correlation was observed between nighttime lights-out time and average value of SpO2 (Avg-SpO2) (β = -0.05, Conclusions: A U-shaped relationship was identified between lights-out time and nocturnal Avg-SpO2 during early pregnancy, with the inflection at 22:00. Notably, later lights-out times are associated with lower levels of Min-SpO2 for advanced maternal age. The findings suggest that appropriately adjusting the duration of light exposure before sleep and maintaining a relatively restful state may be more beneficial for the stability of SpO2 in pregnant women. Conversely, deviations from these practices could potentially lead to pathological alterations in SpO2 levels.