High rate of complications with exchange nailing for femoral nonunions.

OBJECTIVES

To evaluate rates of nonunion repair, osseous healing, and outcomes in femoral nonunions with contemporary healing.

DESIGN

Retrospective review.

SETTING

Five academic level 1 trauma centers.

PATIENTS/

PARTICIPANTS

This study includes adult patients (age older than 18) seen at one of the participating institutions between 2012 and 2019 who sustained a femur fracture (OTA/AO 31, 32, 33) initially treated with intramedullary fixation that developed nonunion and were treated with exchange nailing for the index nonunion surgery. Seventy-nine patients with femoral nonunion met inclusion criteria.

INTERVENTION: Exchange nailing for treatment of femoral nonunion.

MAIN OUTCOME MEASUREMENTS: The primary outcome measure was radiographic osseous union. We further analyzed union rates by OTA/AO classification, nonunion type, implants used, graft used, time from the initial procedure, and infection status.

RESULTS

Seventy-nine patients met inclusion criteria. Rates of osseous union were similar by OTA/AO classification (P = 0.48), nonunion type (hypertrophic, oligotrophic, atrophic) (P = 0.52), implant/biologic used (P = 0.45), and time from the initial procedure until exchange nail procedure (P = 0.09). Forty-two patients had inflammatory laboratory markers (C-reactive protein, erythrocyte sedimentary rate) and cultures obtained during the first nonunion surgery with no significant differences in union (P = 0.29) based on laboratory and culture results. However, a considerable number of complications were encountered (n = 32; 41%). Common complications included reoperation (n = 30; 38%) secondary to recalcitrant nonunion, readmission, implant failure, and infection.

CONCLUSIONS

This large, multicenter study with modern implants, instruments, and techniques for exchange nailing of femoral nonunions demonstrates high rates of reoperation (n = 30; 38%), but higher rates of osseous healing (n = 68; 86% healed) than previously reported data in the literature.

LEVEL OF EVIDENCE: III.