- Objective: The purpose of this study was to evaluate the association between amniotic fluid index (AFI) and adverse perinatal outcome, and whether a critical cutoff can be defined. Methods: A retrospective cohort study was conducted. Included were patients who were admitted to the ultrasound unit of the tertiary medical center between the years 1988 and 2010. Parturients were classified into five groups according to their AFI: <20 (n = 9974; comparison group), 20-23 (n = 2771), 24-27 (n = 1315), 28-31 (n = 494) and 32 + (n = 260). Pregnancy and the perinatal outcomes were compared between the groups. Statistical analysis included the chi-square tests for trends, and multivariable models to control for confounders (with AFI as a dummy variable). Results: A significant linear association was found between AFI and adverse perinatal outcomes including hypertensive disorders, diabetes mellitus, preterm labor, macrosomia, placental abruption and low birth weight. Furthermore, using multivariable logistic regression models, controlling for confounders such as maternal and gestational age, hypertension, diabetes mellitus, etc., the significant association between all four subgroups of AFI > 20 and adverse perinatal outcomes remained. Conclusion: A significant linear association exists between AFI > 20 and perinatal complications such as perinatal mortality, low Apgar scores and preterm labor. Hence, the critical cutoff for polyhydramnios should be re-evaluated.