1 Introduction

Recurrent PFI occurs when the static and dynamic stabilisers of the patellofemoral joint (PFJ) are no longer able to maintain congruency.^1,2 Recurrent patellar dislocation occurs in between 15 % and 60 % of patients,^3,4 with patellofemoral instability accounting for approximately 3 % of knee pathologies.⁴ A number of objective radiological features in recurrent PFI have been identified including trochlear dysplasia, quadriceps dysplasia, patella alta, increased TT-TG distance, increased femoral anteversion and external tibial torsion.⁵ The MPFL is instrumental in providing PFJ stability and is frequently slack or ruptured in patients with recurrent PFI.^6,7

Medial patellofemoral ligament reconstruction has become part of the standard management regime for recurrent PFI.^8,9 Clinical outcomes are varied when this procedure is performed alone and without addressing the underlying anatomical factors which pre-dispose to PFI.^8–10 Increased lateral vector forces caused by anatomical factors which increase the quadriceps angle (Q angle), including excessive femoral anteversion, external tibial torsion and an increased TT-TG, have all been implicated in MPFL-R failure.^10,11

Studies report an incidence of up to 18.7 % of femoral torsional abnormalities in patients with recurrent PFI.¹² A D-DFO can reliably address these abnormalities.^10,12–15 Recent case series have reported favourable outcomes when a D-DFO with lateral plate fixation is performed in combination with a MPFL-R.^7,16

Whilst, retrograde IM nailing potentially offers greater axial strength than a lateral plate,^17,18 there is a paucity of evidence demonstrating D-DFO fixation with IM nails, with only one case report existing in the literature to the authors’ knowledge.¹⁹ This study aimed to determine the early outcomes following D-DFO with an IM nail prosthesis combined with MPFL reconstruction in the treatment of recurrent PFI.

2 Methods

2.3

2.3 Osteotomy

A lateral subvastus approach was used to expose the distal femoral shaft and to facilitate derotation at the osteotomy site. The osteotomy was performed using a transverse bone cut (Fig. 1). An incision medial to the patellar tendon was performed, with a guide wire and reamer passed across the osteotomy under image intensifier guidance (Fig. 1).

Close

Fig. 1 Intraoperative images - Intraoperative images demonstrating guide wire and retrograde nail insertion with DHS wires as levers to de-rotate osteotomy. Also demonstrating intraoperative images of Stryker T2 SCN IM nail with distal locking screws.

Export to PPT

Fixation was performed with a retrograde IM femoral nail (Stryker T2 SCN). A short femoral nail was used in patients with a BMI under 30 kg/m², whereas a long nail was used in patients with a BMI over 30 kg/m². Two 3.2 mm dynamic hip screw wires were used to aid to the correction (Fig. 1). This was facilitated with stab incisions for the two proximal locking screws, followed by either three or four distal locking screws (Fig. 1). The rotation and reduction were checked clinically and radiographically (Fig. 1).

The targeting angle for derotational osteotomy was determined by pre-operative CT rotational profile. If the patient's starting angle was 45°, this was corrected by 20–25°. If their starting angle was 25°, this was corrected by 10–15°. Degree of derotation was measured intra-operatively with a protractor.

2.6

2.6 Radiological parameters

All patients had complete pre-operative anterior-posterior and lateral plain radiographs, axial, sagittal and coronal CT scans and axial, sagittal, and coronal MRI with T1, T2 and Short T1 Image Recovery (STIR) weighted image sequences of the lower limbs to facilitate operative planning. Patients were followed up post-operatively with plain radiographs conducted six weeks following the index procedure to assess fracture healing and the appearance of the construct (Fig. 2). The following radiological parameters were chosen due to their significance in the diagnosis and prognostication of PFI.²²

Close

Fig. 2 Post-operative radiographs - Anterior-posterior (AP) and Lateral X-ray images taken at five months post-operatively illustrating callus formation.

Export to PPT

2.6.1

2.6.1 Tibial tuberosity-trochlear groove

The TT-TG distance in millimetres was measured by drawing lines from the deepest part of the trochlear groove, perpendicular to the posterior condyle. A second line was then drawn through the most anterior portion of the tibial tuberosity, with the distance between the two measured to be the TT-TG distance.

2.6.2

2.6.2 Insall-Salvati ratio

The patella height was determined via the Insall-Salvati ratio. Sagittal MRI scans were utilised to compare the length of the patella tendon to the maximum height of the patella with the knee in extension (Fig. 3).

Close

Fig. 3 Insall-Salvati ratio and Patella tilt angle - T1 weighted sagittal MRI scan demonstrating an increased Insall-Salvati ratio and axial CT scan demonstrating a patella tilt angle >20°.

Export to PPT

2.6.3

2.6.3 Patellar Tilt

Patellar tilt was determined by measuring the angle between the widest axis of the patella and the posterior condylar line on axial CT slices (Fig. 3).

2.6.4

2.6.4 Femoral anteversion angle

All participants had pre-operative FAA measurements on axial CT scans. FAA was defined as the angle between a line drawn connecting the middle of the femoral neck to the centre of the femoral head and the posterior condylar axis (PCA) (Fig. 4).

Close

Fig. 4 Femoral Anteversion Angle - Axial CT scans, demonstrating Femoral Anteversion Angle measurement (FAA°).

Export to PPT

3 Results

4 Discussion

This study demonstrates that D-DFO with IM nail fixation combined with MPFL-R is a potentially viable operative approach in patients recurrent PFI. We report a 0 % post-operative re-dislocation rate in 21 knees with a mean follow-up period of 17 months. This compares with a sample size and follow-up period in recent studies on D-DFO with MPFL-R for recurrent patellofemoral instability in the range of 30–135 knees, with mean or median follow-up periods between 2 and 4 years. Zhang et al. included 49 patients in the DFO + MPFL group and 31 in the control group, with a median follow-up of 4 years.²³ Another large series by Zhang et al. reported 102 knees with a mean follow-up of 4.1 years.²⁴ Other studies have sample sizes ranging from 12 to 66 knees and follow-up periods from 16 months to over 4 years. The meta-analysis by Wang et al., recently published a systematic review of 569 knees, reporting a 1.1 % re-dislocation rate following combined D-DFO and MPFL-R procedures in patients with recurrent patellar dislocation.²⁵ Methods of osteotomy fixation most reported include plate fixation (locking plates or standard plates), with some studies specifying the use of lateral distal femoral locking plates. For example, Zhou et al. describe fixation with lateral locking plates after DDFO.²⁶ Other studies, such as Imhoff et al., also report plate fixation as the standard method.¹³ Bicortical screws may be used in some cases, but plate fixation is the predominant technique in the orthopaedic literature. All studies included utilised distal femoral plate fixation for the de-rotation osteotomy, with no studies presenting findings on concomitant IM nail fixation. We demonstrate comparable re-dislocation rates when utilising an IM nail to fix the osteotomy, albeit, in a smaller cohort of patients. However, IM fixation may carry disadvantages compared to plate fixation. Theoretical risks associated with intra-articular entry during retrograde intramedullary nailing for derotational femoral osteotomy include patellofemoral chondral defects and potential for accelerated patellofemoral osteoarthritis, as the entry point traverses the articular surface and may require joint capsulotomy.^27–30 However, these risks remain theoretical and have not been definitively proven to result in clinically significant impairment or increased complication rates in published clinical studies.^29,30

While the reported complication rate of 14.3 % (n = 3) is marginally greater than previous studies,^13,31–33 one was a case of non-union in a smoker, who reported regular post-operative non-steroidal anti-inflammatory use. This highlights the need for careful patient selection but is not clear whether the complication rate would have been less with plate fixation and restricted post-operative weight bearing seen in other studies.

The evidence base regarding retrograde IM nailing following a D-DFO is limited. The fact that this procedure is seldom performed and reported, hinders the ability to compare its outcomes with other procedures for PFI. Distal femoral osteotomy for patellofemoral instability are most commonly performed using plate fixation, particularly lateral locking plates, which are favoured for their versatility and ability to avoid intra-articular violation. Retrograde intramedullary nailing is less frequently used in this context but is a recognised fixation method for DFO and fractures, especially in extra-articular or partial articular cases notwithstanding the aforementioned theoretical, but not clinically proven, risks associated with intra-articular entry.^34,35 In this study, following osteotomy fixation with a retrograde IM nail, we were able to recommend immediate unrestricted weightbearing for all patients. While one patient in this study experienced non-union, IM nails provide stiffer on axis fixation than lateral plates, reducing strain at the osteotomy, giving a theoretically lower non-union rate.³⁶ It is unclear whether the non-union in this study may have been prevented if weight bearing had been restricted in high risk individuals as in previous studies.¹⁶ Comparisons to literature are challenging as plates are the predominant fixation method for DFO performed for recurrent patellofemoral instability, with intramedullary nails rarely used for this specific indication. The medical literature on patellofemoral instability and derotational femoral osteotomy overwhelmingly reports the use of lateral locking plates, reflecting consensus practice in this patient population.³⁷ When intramedullary nails are used, it is typically in the context of fracture fixation or complex cases, not routine patellofemoral instability surgery.³⁸Complication rates for plate fixation in distal femoral osteotomies—including non-union, malunion, infection, and symptomatic hardware—range from 14 % to 20 %, with reoperation rates for non-union or hardware removal reported at 18.7 % in large series. Implant removal for plates is performed in symptomatic cases or for infection, with removal rates higher than for nails, which are less frequently removed unless symptomatic.^39,40 For intramedullary nails, complication rates such as delayed union, infection, and malalignment are comparable to plates, but reoperation rates for bone healing are lower (3 % for nail-plate constructs vs. 18.7 % for plates alone).³⁹ However, direct data on implant removal rates for nails in the context of patellofemoral instability are lacking, and removal is generally less common than for plates.⁴¹

The mean pre-operative FAA in our cohort was 20.7°. Biomechanical and clinical studies have identified increased FAA as a key factor, necessitating correction in RPI.^10,16 Relative internal rotation of the distal femur increases Q angle, potentiating lateral vector forces on the reconstructed MPFL.^10,16 Recent literature has identified patients with a FAA ranging between 20 and 30° as candidates for a D-DFO.^10,16 A positive J sign is also a strong risk factor for MPFL-R failure.²⁵ Within our study of 21 knees, 38.1 % had a positive pre-operative J sign. This study therefore adds to the growing literature in favour of performing combined osseous and soft tissue procedures in patients with a positive J sign and increased FAA measurements.^7,25,42

The limitations of our work must be considered when interpreting its findings. First, the retrospective nature of the study limits the generalisability of its results. Second, the lack of a comparator group limits our ability to perform direct comparisons between combined D-DFO and IM nail fixation and other procedures for PFI. A larger scale, prospective, comparator study, would aid in this developing field of research. Fourth, we did not include the effect of DFO and IM nail fixation on radiological parameters given there was no clinical indication to do so. Future work may assess post-operative objective radiological indicators, but may require specific ethical approval if ionising radiation is required. Finally, concomitant trochleoplasty procedures in five patients may have affected the post-operative dislocation rates.

5 Conclusion

This study adds to the growing literature base in the management of recurrent patella instability. We report good outcomes following a combined MPFL-R and D-DFO which utilised a retrograde femoral IM nail for osteotomy fixation. Further studies are required to continue to evaluate its feasibility and outcomes.

Informed consent

Not applicable.

Institutional ethical committee approval

Not applicable.

Ethical statement

Not applicable.

Patients’ consent statement

Not applicable.

Author statement

Olufemi Olamide Oladipo Olatigbe: Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – Original draft preparation, Data extraction Diego Agustín Abelleyra Lastoria: Formal analysis, Writing – Original draft preparation, Writing – Review and Editing Matthew Bence: Writing – Original draft preparation, Writing – Review and Editing Carlo Ross: Data extraction, Formal analysis Carlo Ross: Conceptualization, Methodology, Project Administration, Supervision, Writing – Review and Editing Caroline Blanca Hing: Conceptualization, Methodology, Project Administration, Supervision, Writing – Review and Editing.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.