The purpose of this study was to determine the arthroscopic findings in patients with recalcitrant retropatellar pain syndrome (RPPS) and correlate them with the patient's long-term clinical course. All patients undergoing arthroscopy for recalcitrant retropatellar pain syndrome were evaluated. Patients were excluded from the study if there was any history consistent with a meniscal or cruciate injury or if they had previously had knee surgery. Long-term follow-up was obtained in 41 of 81 patients (51%) (range, 24 to 73 months; mean, 51 months). The arthroscopic findings were recorded, and the status of chrondral surfaces graded and correlated with clinical ratings, which utilized a modification of the Insall rating system. Debridement of cartilage irregularities was performed routinely. At follow-up, nearly equal numbers of patients improved, stayed the same, or got worse. No correlation was seen between the findings at arthroscopy and the long-term results. Debridement of cartilage lesions was not found to be beneficial as a treatment modality. Missed intra-articular pathology was found at arthroscopy in only two (4%) of the 81 patients. Based on this study, conservative treatment remains the treatment of choice.