A retrospective follow-up study of a population-based case series was conducted to determine the clinical course and outcomes of carpal tunnel syndrome (CTS). A total of 425 cases first diagnosed between 1979 and 1988 were followed through 1993. Among patients who did not have surgery, median duration of symptoms was between 6 and 9 months, but 22% had symptoms for 8 years or longer. Patients who had surgery were about 6 times more likely to have resolution of their symptoms than were patients who did not have surgery. Patients who had surgery 3 or more years after their initial diagnosis of CTS were less than half as likely to have symptom resolution than were patients who had surgery within 3 years of diagnosis. The results indicate that surgery is a highly effective treatment, but duration of CTS prior to surgery is a key determinant of surgical outcome.