- Research Article
- 10.2106/jbjs.25.01111
Nonunion of Distal Femoral Fractures: Commentary on an article by Robert K. Wagner, MD, et al.: "Risk Factors for Nonunion Following Lateral Locked Plating of Distal Femoral Fractures. A Bayesian Analysis of 560 Patients".
- Feb 04, 2026
- The Journal of bone and joint surgery. American volume
- Carl Haasper
Wagner et al. present an extensive series of distal femoral fractures managed with lateral locked plating, employing Bayesian regression to identify risk factors for nonunion. Their innovative application of posterior probabilities, as opposed to p values, represents an advancement toward a more clinically intuitive interpretation of data. With a cohort of 560 patients across 2 Level-I trauma centers, the study boasts considerable statistical power. Furthermore, the emphasis on a clinically meaningful end point—reoperation to promote union—enhances the study’s practical relevance. The authors examined patient and fracture characteristics as well as surgical variables such as screw density and coronal alignment. They found that varus or valgus malalignment and multifragmentary comminution increased the odds of nonunion; these findings reinforce biomechanical principles and echo the results of prior large cohort studies1,2. Interestingly, their study also demonstrated that a lower screw density may be protective against nonunion, suggesting that overly rigid constructs can suppress callus formation. Clinically, surgeons should prioritize alignment restoration and avoid overly rigid constructs, especially in osteoporotic bone. In complex cases, dual plating or nail-plate constructs may be justified, and primary endoprosthetic replacement is an option for elderly patients with severe comminution1,3. Successful treatment relies on mechanical stability and biological augmentation, with autologous bone grafts being preferred over substitutes and biologics. Overly rigid constructs may hinder bone-healing, whereas insufficient stability may fail to support union, necessitating a balance between mechanical and biological factors. However, there are important caveats to consider. The retrospective design spans nearly 2 decades, during which time implant technologies and surgical techniques have evolved. The study pooled native and periprosthetic fractures despite their biological and technical differences. Relying on return to the operating room as the definition of nonunion may have caused asymptomatic or nonoperatively managed nonunions to be overlooked. Several of the Bayesian credible intervals were broad, reflecting uncertainties despite the large sample sizes. Additionally, although posterior probability thresholds of >95% indicate “very strong evidence,” they essentially parallel traditional frequentist cutoffs and may mislead clinicians who are less familiar with Bayesian terminology. In context, the study reinforces—but does not drastically expand upon—established insights. Earlier cohorts emphasized systemic risk factors such as anemia, open injury, and high body mass index4. The utilization of a Bayesian framework in the study by Wagner et al. is commendable, as such a framework aligns with clinical decision-making by expressing risk in terms of probability rather than binary statements of significance. However, the effect sizes were small to moderate, reminding us that nonunion is multifactorial, and thus optimizing patient biology and surgical technique in tandem is essential. In summary, Wagner et al. deliver valuable, quantitative reinforcement of the importance of alignment and construct strategy in distal femoral fixation. Their Bayesian approach lends clarity and may influence future studies. For now, surgeons should treat these probabilities not as absolutes but as components of a broader operative planning paradigm—one in which precision in reduction and a thoughtful fixation strategy remain foundational.
Read more
2
