Optimal treatment of osteochondritis dissecans of the knee has been controversial. Various treatment methods have resulted in good short-term results. Unfortunately, the age range of the patients who typically require treatment of this disorder is young enough that long-term results become critical. We did a retrospective review to evaluate our hypothesis that simple excision of a loose osteochondritis dissecans fragment results in a higher percentage of unacceptable outcomes when compared with treatment methods that preserve the articular cartilage. Between 1982 and 1993, 30 knees in 29 patients had arthroscopic or arthroscopically assisted surgical excision of a symptomatic loose osteochondritis dissecans lesion of the knee. Seventeen patients who had excision were available for followup. There were 12 males and five females. The average age of the patients at the time of surgery was 26 years (range, 12.5-38 years). The average length of followup was 8.9 years (range, 4-15 years). Results were graded using the Hughston rating scale for osteochondritis dissecans. Only six of 17 patients (35%) had a good or excellent result. Eleven of 17 (65%) had a fair or poor result. We think our results at intermediate followup support our hypothesis, and we recommend aggressive attempts to preserve the articular cartilage and avoid excision of the fragments when possible.