Background: Methotrexate (MTX) discontinuation is recommended for women before pregnancy because MTX, a first-line treatment rheumatoid arthritis (RA), can be risk of spontaneous abortion (SAB). However, the optimal washout period required for the SAB risk to return to baseline levels is uncertain. The aim of this study was to estimate the timing of MTX discontinuation before the last menstrual period (LMP) at which the SAB risk becomes comparable to unexposed women. Methods: This nationwide cohort study utilized South Korea’s claims database (2015–2024) to identify pregnancies among women with RA. The exposed group was pregnancies with MTX use within 180 days before LMP; pregnancies without MTX use during this period were classified as unexposed. The outcome was SAB, defined as fetal death within 20 weeks’ gestation. Induced abortion before 20 weeks was treated as a competing pregnancy outcome. To address potential informative censoring due to MTX re-prescription after the LMP, we censored follow-up at re-prescription and applied inverse probability of censoring weighting (IPCW) based on baseline age and clinical characteristics. The baseline risk was defined as the weighted probability of the outcome in the unexposed group. We then used IPCW-weighted logistic regression with restricted cubic splines to model the nonlinear association between the interval from the end date of the last MTX prescription to the LMP and SAB risk, adjusting for covariates, and estimated predicted SAB risks with 95% confidence intervals (CIs). Results: Of 11,880 pregnancies in the overall study population, 605 (5.1%) were MTX-exposed. The baseline risk of SAB in the non-exposed group was 20.5% (95% CI, 19.8–21.3%). The risk peaked at 44.5% (95% CI, 38.5–50.8%) when the interval between the last prescription and LMP was 34 days in the exposed group. However, the risk declined and converged to the baseline level (20.5%) at 110 days. Conclusion: In women with RA, MTX exposure near LMP was associated with a significant increase in SAB risk, which persisted until 110 days after the last day of supply. These findings are consistent with guideline recommendations of at least a 3-month washout period and suggest that a 4-month interval after the last MTX supply may offer additional reassurance. This potential benefit should be balanced against the need for disease control.