Patellar instability most frequently presents during adolescence. Congenital and infantile dislocation of the patella is a distinct entity from adolescent instability and measurable abnormalities may be present at birth. In the normal patellofemoral joint an increase in quadriceps angle and patellar height are matched by an increase in trochlear depth as the joint matures. Adolescent instability may herald a lifelong condition leading to chronic disability and arthritis. Restoring normal anatomy by trochleoplasty, tibial tubercle transfer or medial patellofemoral ligament (MPFL) reconstruction in the young adult prevents further instability. Although these techniques are proven in the young adult, they may cause growth arrest and deformity where the physis is open. A vigorous non-operative strategy may permit delay of surgery until growth is complete. Where non-operative treatment has failed a modified MPFL reconstruction may be performed to maintain stability until physeal closure permits anatomical reconstruction. If significant growth remains an extraosseous reconstruction of the MPFL may impart the lowest risk to the physis. If minor growth remains image intensifier guided placement of femoral intraosseous fixation may impart a small, but acceptable, risk to the physis. This paper presents and discusses the literature relating to adolescent instability and provides a framework for management of these patients. Cite this article: Bone Joint J 2017;99-B:159–70

This study was performed to evaluate the results of transphyseal anterior cruciate ligament (ACL) reconstruction in children with open physes. Between 2001 and 2009, 31 knees in 29 patients with a mean age of 14.1 years (10-16) underwent transphyseal, arthroscopic ACL reconstruction using an autogenous four-strand hamstrings graft. The technique was identical to that used in the adult population in our unit, except that care was taken to ensure fixation did not cross the physes. The patients were followed up to the point of physeal closure at skeletal maturity. The mean length of follow up was 25 months. The primary outcome measure was graft survival. Functional outcome was measured using Lysolm scores, International Knee Documentation Committee (IKDC) scores and the Tegner activity scale. Post-operative clinical examination and radiological findings were also recorded. There were two cases of re-rupture following a further sports injury, one 12 months post-operative and one at six years. Both patients later had successful revision ACL reconstructive surgery. The re-rupture rate in this case series was 6%. Two patients underwent repair of a concomitant lateral meniscal tear at the time of surgery. The mean post-operative Lysolm score was 88 and the mean IKDC score was 88.1. The mean Tegner activity scale was 7.95. There was no evidence of growth disturbance radiologically or leg length inequality clinically. There was one case of deep infection which was successfully treated with two subsequent arthroscopic washouts. This study demonstrates that transphyseal arthroscopic ACL reconstruction using hamstrings graft and an ‘adult’ technique, in skeletally immature patients, provides good functional outcomes, has a low revision rate and is not associated with physeal growth arrest or disturbance