OBJECTIVE Our objective was to evaluate the effect on the neonatal thyroid function of povidone-iodine (PVP-I) used on mothers during the perinatal period. METHODS Eight consecutive-term pregnant women were randomly separated into 4 groups. Either PVP-1 or benzethonium chloride (BC) was used to disinfect both maternal skin during labor and vaginal lacerations after delivery in Group I and IV; and PVP-1 or BC was supplied to either skin or lacerations in Groups II and III. On day 5, we measured the iodide concentration in breast milk and random urine of neonates, and the levels of neonatal thyroid-stimulating hormone (TSH) and free thyroxine. RESULTS The TSH levels in Group I and the breast-milk iodide concentrations in Groups I and II were significantly higher than those in Group IV (p < 0.05). Dietary iodide intake from breast milk correlated significantly with urinary iodide concentrations (r = 0.52, p < 0.001). In our clinic, the recall rate at the time of screening for congenital hypothyroidism was significantly decreased by discontinuing the use of PVP-I (4.47% vs 0.74%, p < 0.001). CONCLUSIONS These findings suggest that the use of PVP-I on mothers is associated with transient neonatal hyperthyrotropinemia, in which breast milk might play an important role. Therefore, we recommend that the use of PVP-I be avoided during labor.