A prospective, descriptive study of leptospirosis in Aquitaine was conducted from 2004 to 2006 to improve case identification and characterization, detect and investigate clusters of cases, and assess the effectiveness of mandatory reporting. A case was defined as any person residing in one of the five departments of Aquitaine between 2004 and 2006 who presented with clinical signs suggestive of leptospirosis. Based on serological results, cases were classified as confirmed, probable, or suspected. In the first year, a passive data collection method was tested. Starting in the second year, surveillance became semi-active. From 2004 to 2006, 54 cases of leptospirosis were recorded, including 36 confirmed cases and 18 probable cases. Overall, incidence rates were low and had decreased compared to previous years. Nevertheless, in three departments—and particularly in the Dordogne—the incidence rate was significantly higher than that of metropolitan France. The "inactive" population was the most represented, and among the active population, several cases were engaged in occupations with a risk of exposure. Direct or indirect contact with rats was the most common risk factor. Activities involving contact with freshwater without immersion were generally more common. More than 87% of cases required hospitalization, and 22% were treated in an intensive care unit. Serogroup Ictérohaemorrhagiae was predominant, followed by Grippotyphosa. This study demonstrated that mandatory reporting of leptospirosis is currently unnecessary due to the existence of a National Reference Center that already conducts effective surveillance tailored to public health concerns. Furthermore, the complex, nonspecific case definition—based primarily on laboratory diagnosis—leads to long response times. However, given the severity of the disease and its potential for progression, surveillance efforts could be strengthened in departments with high incidence rates. (R.A.)