Dislocations of the patella have been reported to occur more often in children and adolescents (Buchner et al. 2005
). For first-time patella dislocations, nonoperative management is recommended with early joint motion and quadriceps strengthening after initial long leg casting or bracing (Sillanpaa et al. 2009
). Recurrent patella dislocations that fail to respond to nonoperative treatment require operative intervention to prevent chondral damage and subsequent osteoarthritis (Maenpaa et al. 1997a
, Barber and McGarry 2008
In young patients with open physes, surgery in the area of the growth plates could result in premature closure of the physis (Grammont et al. 1985
, Nelitz et al. 2011
). There have been few studies addressing the long-term outcome of operative treatment of recurrent patella dislocation in immature patients.
In 2009, one of the authors (FS) presented preliminary results of the modified Grammont technique after a mean of 4 years (Schneider and Linhart 2009
). In that study, the redislocation rate was 4 out of 36 and the mean Tegner activity score declined to 1.5 points. 7 patients in the present study also took part in the earlier study by Schneider et al. (12 knees) (Schneiderand Linhart 2009
). In contrast to that study, we applied stricter inclusion criteria in the present study (see Material and methods).
Several procedures for the operative treatment of recurrent patella dislocations have been reported (). However, due to the variability of measured parameters, meaningful comparisons between the studies mentioned are not possible. Several studies included adults only, while other studies included a small number of patients. The optimal management of patellofemoral instability in children and adolescents therefore remains unclear. Recent studies have identified the medial patella-femoral ligament (MPFL) as the strongest passive stabilizer of the patella in preventing excessive lateralization (Beasley and Vidal 2004
). Reconstruction of the medial femuropatellar ligament (MPFL) is becoming more and more popular (Beasley and Vidal 2004
). The femoral approach for the MPFL is close to the distal physis of the femur, with a risk of growth disturbances (Nelitz et al. 2011
). Previous studies addressing the outcome of an MPFL reconstruction in immature patients are limited in number, with only short follow-up periods. They may therefore have under-reported potential growth disturbances following surgery (Schottle et al. 2005
, Yercan et al. 2011
In more than 80% of patients with open physes and recurrent patella dislocations, there are predisposing factors (Dejour et al. 1994
, Maenpaa 1996). In particular, torsional abnormalities appear to be more frequent than commonly appreciated. Increased femoral antetorsion in combination with increased lateral tibial torsion might favor patella dislocations in young patients (Airanow et al. 1990
). 50 of our patients showed increased internal rotation of the femur and 16 showed increased external rotation of the tibia. These conditions are directly addressed by the modified Grammont procedure. In patients with closed physes who have an abnormal lateral position of the tibial tuberosity, distal bony procedures provide an appropriate alternative, with established success rates (Endres and Wilke 2011
). Open apophysis of the tibial tuberosity contraindicates osteotomy at the tibial tubercle, as it has been reported to lead to closure of the physis and genu recurvatum (Grammont 1985
, Barber and McGarry 2008
). The modified Grammont procedure we used—an exclusively soft-tissue technique—restores the distal re-alignment and preserves the tibial apophysis. In the present study, no growth disturbance occurred. This simple technique provides little operative trauma, allows integrity of the pes anserinus tendons, and gives dynamic positioning of the distal patella tendon. The pre- to postoperative unchanged Blackburne-Peel index indicated that the surgery had no effect on the length of the patella ligament and only medialized it. Early weight bearing is possible, and early movement is even indicated. In contrast, bony distal re-alignment procedures require longer rehabilitation and bear the risk of non-union.
It is well known that patients who undergo stabilizing surgery have a higher incidence of osteoarthritis over time. Arnbjornsson et al. (1992)
reported a 75% rate of osteoarthritis at a mean follow-up of 14 years. Sillanpaa et al. reported a severe osteoarthritis rate of 30% after 10 years (Sillanpaa et al. 2011
A patellar ligament that is fixed too far medially leads to increased medial articular cartilage contact pressure, resulting in an early medial compartment arthritis, especially in patients with genua valga (Kuroda et al. 2001
). We modified the original Grammont technique by allowing dynamic positioning of the distal patella tendon. We expected that the patella would find its “ideal” position, reducing deforming forces and minimizing the risk of chondral damage, which may explain the comparably low rate of osteoarthritis (6/58) in our patients.
In the present study, 54 of 65 knees had no more dislocations. None of the 8 patients with a single redislocation required revision surgery. Redislocations occurred in 2 knees with type-C dysplasia according to Dejour et al. (Dejour et al. 1994
). For these patients, the simple patellar re-alignment does not appear to be sufficient, and as adults these patients were offered MPFL reconstruction. Notably, more than half of the patients in our study showed a patello-femoral dysplasia, which plays a considerable role in recurrent dislocations (Nelitz et al. 2012
). Furthermore, the young age of the patients appears to favor recurrent dislocation (Buchner et al. 2005
The average Lysholm score in the present study was 82 points, revealing good to very good functional outcome. Redislocations do not appear to worsen clinical function, which is in accordance with the findings of Buchner et al. (2005)
. Other authors (Maenpaa et al. 1997b
, Nikku et al. 1997
) also found the discrepancy between a high recurrence rate and good subjective outcome. Buchner et al. (2005)
reported a redislocation rate of 10/37 (surgical group) after acute patellar dislocation, with a good subjective outcome in 27/30.
The strengths of our study are the length of follow-up, the low dropout rate, and a uniform technique. Weaknesses of the study include the missing preoperative Lysholm scores for comparison, the lack of a control group, the lack of data on Q-angles, and the lack of additional radiographic measurements.
In summary, we found that a modified Grammont procedure is a feasible method to treat recurrent patella dislocation in skeletally immature patients. However, patients with higher-grade patello-femoral dysplasias (Dejour type C) should be informed that they have a higher risk of redislocation and that additional surgery might be required after skeletal maturity.