Objective To explore the effectiveness of arthroscopic medial retinaculum pl ication (MRP) for recurrent patellar dislocation in adolescents. Methods Between March 2000 and October 2007, 30 adolescent patients with recurrent patellar dislocation underwent arthroscopic MRP, and 28 of them (12 left knees and 16 right knees) completed thefinal follow-up and were enrolled. There were 5 males and 23 females with an average age of 14.7 years (range, 12-19 years). The dislocation duration was 4 to 39 months with an average of 18.8 months. All patients experienced 2-4 episodes of dislocation. Before operation, all the patients showed positive apprehension test; the degree of lateral patellar translation was (2.9 ± 0.7)° and there was no hardness termination in lateral translation. The International Knee Documentation Committee (IKDC), Lysholm, Kujala, and Tegner scores were 47.7 ± 3.7, 52.6 ± 4.9, 66.7 ± 5.9, and 3.1 ± 1.3, respectively. All patients underwent arthroscopic MRP procedure. Results All incisions healed by first intention without compl ication. All the patients were followed up 2-7 years (4.8 years on average). During the follow-up, more and more patients showed positive apprehension test, and at 24 months of follow-up, 12 patients showed positive apprehension test; less and less patients had hardness termination in lateral translation, only 3 patients at 24 months of follow-up; the degree of lateral patellar translation increased, (2.3 ± 1.1)° at 24 months of follow-up. There were significant differences in positive apprehension test and hardness termination between preoperation and 24 months of follow-up (P lt; 0.05). Moreover, 6 patients suffered from redislocation, and 23 patients experienced patella instabil ity. The IKDC, Lysholm, Kujala, and Tegner scores at 24 months of follow-up were 62.5 ± 6.2, 70.7 ± 5.1, 76.6 ± 4.8, and 3.9 ± 0.7, respectively, showing significant differences when compared with preoperative scores (P lt; 0.05). CT examination showed that the congruence angle, lateral patellar angle, and patellar tilt angle were recovered to normal level after operation immediately, however, the final patellar position at 24 months of follow-up was not significantly betterthan that before surgery (P gt; 0.05). There was significant difference in the lateral patellar displacement between preoperation and 24 months of follow-up (P lt; 0.05). Conclusion Arthroscopic MRP is not rel iable for maintaining the corrected position of the patella for recurrent patellar dislocation in adolescents, though functional improvements are significant.