What Is the Likelihood of Union and Frequency of Complications After Parallel Plating and Supplemental Bone Grafting for Resistant Distal Femoral Nonunions?

Abstract

Background: Management of resistant distal femur nonunions is challenging because patients not only have disability from an unhealed fracture, but also often have a shortened femur, stiff knee, deformities, and bone defects to address during revision surgery. Dual plating of the distal femur in such a setting can maintain stability that allows the nonunion to heal while also addressing bone defects and correcting deformities simultaneously. Dual-plating techniques that have been described lack standardization with regard to the size and type of medial-side implants and configuration of the dual-plate construct.

Questions/purposes: (1) What proportion of patients achieve radiologic evidence of union after parallel plating of resistant distal femoral nonunions? (2) What improvements in function are achieved with this approach, as assessed by improvements in femoral length discrepancy, knee flexion, and patient-reported outcome scores? (3) What complications are associated with the technique?

Methods: Between 2017 and 2020, the senior author of this study treated 38 patients with resistant distal femoral nonunions, defined here as nonunions that persisted for more than 12 months since the injury despite a minimum of two previous internal fixation procedures. During the study period, our preferred technique for treating aseptic, resistant distal femoral nonunions was to use dual plates in a parallel configuration augmented with autografts. Of 38 patients, three patients with active signs of infection who underwent resection and reconstruction using bone transport techniques and two patients older than 65 years with deficient distal femur bone stock who underwent endoprosthetic reconstruction were excluded. Of the 33 included patients, 67% (22 of 33) were male. The median age was 40 years (range 20 to 67 years). Nonunion was articular and metaphyseal in 13 patients and metaphyseal only in 20 patients. Our surgical approach was to remove existing implants, perform intraoperative culturing to rule out infection, debride the nonunion, correct the deformity, perform intra-articular and extra-articular lysis of adhesions with quadriceps release, and apply fixation using medial and lateral fixed-angle anatomic locked implants positioned in a parallel configuration. Every attempt to improve length was undertaken, and the defects were filled with autografts. A total of 97% of patients were followed until union occurred (one of 33 was lost to follow-up before union was documented), and 79% (26 of 33) were assessed for functional outcomes at a minimum of 2 years (median 38 months [range 25 to 60 months]) after excluding patients lost to follow-up and those in whom union did not occur after parallel plating. Union was defined as evidence of central trabecular bridging on AP radiographs and posterior cortical bridging on lateral radiographs. These radiologic criteria were defined to overcome difficulties in assessing radiologic healing in patients with lateral and medial plates. With parallel plating, bridging trabecular bone along the posterior cortex on lateral radiographs and the central region on AP radiographs is visualized and can be appreciated and interpreted as evidence of healing in two orthogonal planes. Preoperative and follow-up clinical assessment of knee ROM, the extent of femoral length correction based on calibrated femoral radiographs before and after surgery, and the evaluation of improvement in lower limb function based on the preoperative and follow-up differences in responses to the lower extremity functional scale (LEFS) were studied (the LEFS is scored from 0 to 80, with higher scores representing better function). Complications and secondary surgical procedures to address them were abstracted from a longitudinally maintained trauma database.

Results: Sixty-seven percent (22 of 33) of nonunions showed radiologic healing by 24 weeks, and another 24% (eight of 33) healed by 36 weeks. Six percent (two of 33) did not unite, and one patient was lost to follow-up before union was documented. In the 79% (26 of 33) of patients available for final functional outcome assessment, the median femoral shortening had improved from 2.4 cm (range 0 to 4 cm) to 1.1 cm (range 0 to 2.3 cm; p < 0.001), and the median knee ROM had improved from 70° (range 20° to 110°) to 100° (range 50° to 130°; p = 0.002) after surgery. The median LEFS score improved to 63 (range 41 to 78) compared with 22 (range 15 to 33; p < 0.001) before surgery. Serious complications, including major thromboembolic events, iliac graft site infection, knee stiffness (flexion < 60°), and medial plate impingement necessitating removal, were seen in 30% (10 of 33) of patients. Secondary surgical interventions were performed in 24% (eight of 33) of patients to address procedure-related complications.

Conclusion: Based on our findings, a high likelihood of union and improvements in knee and lower limb function can be expected with parallel plating of resistant distal femur nonunions using anatomic locked plates. However, the increased frequency of complications observed in our study suggests the need for improvements in dual-plating techniques and to explore possible alternative fixation methods through larger multicenter comparative studies.

Level of evidence: Level IV, therapeutic study.