Introduction

Lateral patellar dislocation is a common knee injury in the skeletally immature adolescent [6, 8, 9, 19, 22]. The dislocation causes a traumatic disruption of the previously uninjured medial patellofemoral ligament (MPFL), a part of the medial parapatellar structures [19]. The stabilizing role of the MPFL as the main restraining force to lateral displacement of the patella has been emphasized by many authors [1, 3, 4]. The importance of an anatomical repair respecting the femoral and patellar insertion of the ligament has been proven by several authors [1, 10, 14, 15, 21]. In a cadaver study, Baldwin showed that the femoral insertion of the ligament has a mean width of 10.6 ± 2.9 mm in adults and that the MPFL additionally has an oblique decussation originating from the superficial medial collateral ligament [2]. Schöttle et al. [16] and Redfern et al. [17] described radiographic landmarks to determine the femoral insertion of the MPFL. These landmarks can be relied upon at surgery for the operative ligament reconstruction. Due to the adjacent physis, the management of patellofemoral instability in children and adolescents is challenging. Reconstruction of the MPFL has been advocated as the treatment of choice in skeletally immature patients as well [4, 8, 22].

The literature has conflicting evidence about the relationship of the femoral insertion of the MPFL to the distal femoral physis. Using the indirect radiographic method described by Schöttle et al. [17], Shea et al. [18] found the MPFL origin to be just proximal to the femoral physis in boys and girls that approached skeletal maturity. In contrast, Ladd et al. [12] found the femoral insertion on MR imaging to be below the distal femoral physis. In another study by Kepler et al. [11], the anatomical insertion of the MPFL as seen on MRI scan was distal to the physis in 86% of their patients.

Thus, the relation of the femoral origin of the MPFL to the distal femoral physis in growing children is not yet known. As a result of these conflicting results, the purpose of this study was to analyse true lateral radiographs of children and adolescents with open growth plates to determine the anatomical relations of the origin of the MPFL and the distal femoral physis considering its complex anatomy. This has important impact on the anatomical reconstruction of the MPFL in children and adolescents.

Materials and methods

Antero-posterior and true lateral radiographs from PACS (picture archiving and communication system) computer records of 27 patients with a history of patellofemoral instability were assessed. The mean age of the 27 patients with open growth plates (10 boys and 17 girls) was 14.3 years (range 12–16 years). Images were obtained with the patient in supine position and partial knee flexion. Inclusion criterion was a perfect superimposition of the posterior femoral condyles. On every radiograph, a scale was placed allowing the correction of magnification. All measurements were made on the PACS imaging software to the tenth of a millimetre.

To determine the femoral origin of the MPFL, the method by Schöttle et al. [16] and the method by Redfern et al. [17] were applied independently on the lateral radiograph (Fig. 1).The projected MPFL origin–physis distance was obtained by a perpendicular line from the origin of the MPFL to the physis.

Fig. 1
figure 1

MPFL insertion according to Schöttle et al. (red dot) and Redfern et al. (black dot). Relationship of the femoral physis to the insertion of the MPFL. In this case, the line through the axis of the apex of the Blumensaat line is on the lateral projection of the distal femoral physis

Full size image

On the anteroposterior radiograph, the distance between the most medial part of the physis and a line tangential to the central part of the physis (the physeal line) was measured to quantify the concave curvature when viewed from a superior aspect of the physis (Fig. 2).

Fig. 2
figure 2

On the ap radiograph, the distance between the medial most part of the physis (proximal line) and a line parallel to the central part of the physis was measured to quantify the extent of the concave curvature

Full size image

To cross-reference the femoral insertion of the MPFL onto an AP view, the projected MPFL origin–physis distance was subtracted from the distance between the most medial part of the physis and the central part of the physis.

Both antero-posterior radiographs and true lateral radiographs were assessed by two independent raters at two different time points at least 3 months apart. Intra- and interrater agreement was investigated using the intraclass correlation coefficient (ICC). For calculation of the ICC mixed effects, regression models were fitted using SAS software, version 9.2 (SAS Institute, Cary, North Carolina).

Results

The projected median origin of the MPFL as measured on a lateral radiograph was located 3.2 mm (1.2–5.8 mm) proximal to the physis. As the MPFL insertion according to Schöttle et al. and Redfern et al. only differed in the coronal plane and no difference was found in the transversal plane, no difference in the measured distance between the method by Schöttle et al. and the method by Redfern et al. could be seen [16, 17].

The median distance between the most medial part of the physis and the physeal line on the antero-posterior radiograph was 9.9 mm (4.1–12.0 mm).

Subtracting the two measured values, the median origin of the MPFL was found to be 6.4 mm (2.9–8.5 mm) distal of the physis in all patients.

The overall intraobserver intraclass correlation coefficient was 0.964; the overall interobserver intraclass correlation coefficient was 0.978. This indicates that raters were able to obtain highly reproducible measurements.

Discussion

The findings of the present study showed that the femoral origin of the MPFL as measured on radiographs is distal to the medial femoral physis. This information is of paramount importance, as MPFL reconstruction has become a widely accepted surgical procedure in the treatment of patellofemoral instability in children and adolescents.

Many authors emphasized the need of an anatomical reconstruction with tightening of the MPFL in knee flexion when the femoral origin is placed too proximally [1, 3, 10, 14, 15, 20, 21]. Camp et al. [5] found the failure to restore the anatomical femoral insertion to be a main risk factor for the failure of MPFL reconstruction.

For adults, radiographic landmarks for reconstruction were described by Schöttle et al. [17] and Redfern et al. [16].

Shea et al. [18] tried to determine the origin of the MPFL in relation to the physis by the use of the criteria described by Schöttle et al. [17] and showed that the femoral origin of the MPFL is typically 2–5 mm proximal to the femoral physis.

The results of the present study whose mean age of patients is similar to the mean age of the patients in the study by Shea et al. [18] confirmed their results. Using the criteria by Schöttle et al. [17] and Redfern et al. [16] on the lateral view, the median projected MPFL origin–physis distance in our study was 3.2 mm (1.2–5.8). But it has to be taken into consideration that the lateral view only shows the central part of the physis and that the distal femoral growth plate has a complex anatomy (Fig. 3). In an MRI study, Craig et al. [7] demonstrated that the distal femoral physis has a major curvature inferiorly, being concave when viewed from a superior aspect. In the present study, the median distance from the central part of the physis to the most medial part of it was 9.9 mm. Hence, due to the curvature of the medial part of the physis, the cross-reference onto an AP view showed that the same point that is projected on or proximal to the physis on the lateral view is distal to the physis on the AP view (Fig. 4).

Fig. 3
figure 3

The MRI of a 16-year-old girl illustrates the major concave curvature of the medial distal femoral physis

Full size image
Fig. 4
figure 4

Intraoperative cross-reference of the physis on the lateral view onto an ap view reveals the distance between the medial most part of the physis and the central part of the physis as seen on the lateral view

Full size image

In all patients, the femoral insertion point was distal to the femoral physis (median 6.4 mm) although the lateral radiographic projection showed a location on the physis or proximal of it. As Shea et al. [18] did not consider the concave curvature of the physis, their conclusion was different from ours.

On MRI scan, Kepler et al. [11] measured the distance between the MPFL insertion onto the distal femur and the medial distal femoral growth plate or physeal scar.

The femoral MPFL insertion averaged 5 mm distal to the femoral growth plate (range 7.5 mm proximal to 16 mm distal).

Cadaver studies by LaPrade et el. [13] and Baldwin et al. [2] also confirm the results of the present study that the femoral insertion of the MPFL is distal to the femoral physis.

In summary, the results of this study confirm the assumption that the femoral insertion of the MPFL is distal to the physis. Due to the concave curvature of the distal femoral physis, the lateral radiograph can be misleading for the determination of the relation between the distal femoral physis and the medial patellofemoral ligament. These results should be considered during the repair or reconstruction of the MPFL in skeletally immature patients.

This study has several limitations. In younger children with open growth plates, radiographic landmarks become more difficult to define. The posterior point of the Blumensaat line comes close to the physis and is not readily identifiable. Although our results are in consensus with the available MRI studies, the used technique only gives approximate values. An anatomical study of skeletally immature subjects would be preferable to define the precise anatomical insertion of the MPFL.

Conclusion

Applying the results of this study as well as those described in the literature of the radiographic landmarks to determine the femoral insertion of the MPFL, it can be assumed that the insertion is typically distal to the femoral physis. As a too proximal insertion of the graft can cause unintentional tightening of the MPFL in knee flexion and overload of the medial patellar facet, these results have to be considered when performing reconstruction of the MPFL in children and adolescents with open growth plates.