Surveillance of acute flaccid paralysis (AFP) is the golden strategy recommended by the WHO to verify the condition of polio eradication in a country. Because of the difficulty to detect all of the expected AFP cases and to reach the target incidence of 1/100,000 requested by WHO, the surveillance of enteroviruses in the population has been adopted by several countries as an important additional method to verify the absence of wild-poliovirus circulation. To complete the results of AFP surveillance set up in Italy in 1996, we have conducted a wild poliovirus surveillance by examining stool samples from 1551 healthy children aged less than 5 years, collected during the period January 1997 to January 1998. The children were from three cities (Parma, Rome and Bari) located in northern, central and southern Italy. Thirty-nine polioviruses, 72 non-polio enteroviruses and 50 enteric, nonentero (NE), viruses were isolated from stool specimens. Polioviruses identified were nine type 1, seven type 2 and twenty-three type 3. Characterization of isolates by both antigenic and molecular methods showed that all polioviruses were of vaccine origin. As expected, most polioviruses, especially types 2 and 3, presented retromutations known to be associated with loss of the Sabin attenuated phenotype. The results of this study support the data obtained from the active AFP surveillance conducted in Italy in the same period--on the absence of paralytic disease due to wild poliovirus--and altogether demonstrate the effectiveness of the vaccination program.