Objective To discover the relationship between the post-operative positive residual patellar J sign and the laxity of reconstructed medial patellofemoral ligament (MPFL) after MPFL reconstruction combined with tibial tubercle osteotomy (TTO). Methods A total of 39 consecutive recurrent patellar dislocation (RPD) adult patients who were performed MPFL reconstruction and TTO with more than 2 follow-up time were analyzed retrospectively in the present study. The patellar tracking of all the patients was evaluated and was classified into positive patellar J sign group and negative J sign control group during knee active flexion and extension at final follow up. The computed tomography (CT) examination was performed in all patients at 0° extension of the knee. The patellar laxity index measured by patellar glide test (PGT) and the radiographic parameters (tibial tuberosity-trochlear groove distance, bisect offset index, patella trochlear-groove distance, patella lateral tilt angle, femoral anteversion angle, tibial external angle and knee rotational angle) calculated by CT scans slices as well as the pre-/post- operative knee functional scores including International Knee Documentation Committee (IKDC) score, Kujala score, and Lysholm score were compared between the positive group and negative group at final follow up. Results The average follow-up duration was 26.3±6.7 months (range 24-31) and all the patients did not suffer from the RPD during the follow-up. Ten (26%, 10/39) patients performed positive J sign, and the remaining 29 (74%, 29/39) performed negatively. The two groups were compatible with no significant difference in age, gender, injury side, follow-up duration and preoperative knee function scores (P>0.05). At the final follow up, the patellar laxity index in the positive J sign group was 34.9%±6.9%, while that in the negative group was 24.6%±8.6%. There was significantly difference in the patellar laxity index between two groups (t=-3.413, P=0.002). The bisect offset index in the positive group was 102.9%±12.4%, while that in the negative group was 76.0%±18.4%. There was significantly difference in the bisect offset index between two groups (t=-4.268, P=0.000). The patella trochlear-groove distance in the positive group was 2.2±3.7 mm, while that in the negative group was -7.6±5.8 mm. There was significantly difference in the patella trochlear-groove distance between two groups (t=-4.233, P=0.000). The two groups were compatible with no significant difference in tibial tuberosity-trochlear groove distance, patella lateral tilt angle, femoral anteversion angle, tibial external angle and knee rotational angle (P>0.05). The IKDC, Kujala, and Lysholm in the positive group were 86.4±6.5, 85.3±1.1, 91.5±7.9 respectively with significantly differences compared with preoperative scores 53.3±2.4, 73.2±9.7, 76.7±6.6 (t=-10.163, P=0.000; t=-1.274, P=0.023; t=-8.533, P=0.018). The IKDC, Kujala, and Lysholm in the negative group were 87.3±8.6, 84.3±4.7, 93.1±4.3 respectively with significantly differences compared with preoperative scores 49.1±4.6, 71.4±6.5, 79.5±7.0 (t=-14.381, P=0.000; t=-0.068, P=0.037; t=-6.902, P=0.013). There were no significant difference in the postoperative knee function scores between the two groups (P>0.05). Conclusion RPD patients can recover the patellar instability from the surgical technique of MPFL reconstruction and TTO and none of them suffer from the RPD during the follow-up. However, the morbidity of post-operative positive residual patellar J sign was 26%. The positive J sign might increase the laxity of patella. Key words: Patellofemoral joint; Wounds and injuries; Arthroscopy; Joint instability