JSES - 2026-06-01 - Journal Article
Predictors for outlier results after radial head replacement for acute complex elbow instability.
Diez Sánchez B, Palacios-Díaz L, Antuña SA, Barco R
Topics
Key Takeaway
Female sex (OR 20.29) and younger age (OR 0.94 per year) are independent predictors of outlier poor outcomes (OES <35) after radial head replacement for acute complex elbow instability.
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study identified predictors of poor functional outcomes (OES below p25, <35 points) in 134 patients undergoing RHR for acute complex elbow instability, comparing the bottom quartile (n=35) to the top quartile. On multivariate analysis, younger age (OR 0.94, 95% CI 0.88–0.99) and female sex (OR 20.29, 95% CI 2.27–181.8) were the only independent predictors of outlier poor outcomes. All 10 implant removals occurred exclusively in the p25 group, and terrible triad and Essex-Lopresti injuries were overrepresented in poor-outcome patients.
Key Limitation
The OR for female sex (20.29) is statistically unstable given only 23 women in the p25 group and a 95% CI spanning nearly two orders of magnitude, making this the most clinically cited finding also the least reliable.
Original Abstract
BACKGROUND
Radial head replacement (RHR) is a well-established treatment for comminuted fractures (radial head fracture). In most cases, it provides satisfactory long-term functional results. However, some patients have pain and significant functional limitations. Our aim was to identify patients who underwent an RHR for acute trauma and had a clinical outcome that was considered an outlier. We assessed specific factors that contributed to an unfavorable result.
MATERIALS AND METHODS
From a cohort of 134 patients undergoing RHR for acute complex elbow instability, the Oxford Elbow Score (OES) was used to identify patients whose score was below the p25 (<35 points). Thirty-five patients (23 women and 12 men) with a mean age of 48 years (18-78) and a mean follow-up of 77 months (24-192) were below the p25. The univariate analysis included preoperative demographic characteristics, implant design, diagnosis, and postoperative clinical and radiographic outcomes. Results were compared to patients above the p75. An additional multivariate analysis was performed to identify predictive variables of a p25 OES.
RESULTS
Patients belonging to p25 OES were significantly younger (48 vs. 58 years, P = .02), had a higher energy trauma (P = .01), and their diagnosis was more commonly a terrible triad (18 vs. 11, P = .004) or an Essex-Lopresti (5 vs. 0) injury. Twenty-two patients received a smooth-stem RHR and 13 an anatomic design, with no correlation between implant design and p25 OES (P = .83). The p25 OES showed significantly lower flexion-extension (P = .01), pronation (P = .01), Visual Analog Score (VAS) satisfaction (P < .001), and Mayo Elbow Performance Score (MEPS), and Disabilities of the Arm, Shoulder, and Hand (DASH) scores (P < .001), with more pain (P < .001). All patients who required implant removal belonged to the p25 OES (10 vs. 0). On multivariate analysis, odds of belonging to p25 OES were greater only for younger (odds ratio = 0.94, 95% confidence interval: 0.88-0.99, P = .04) and female patients (odds ratio = 20.29, 95% confidence interval: 2.27-181.8, P = .007). A very strong correlation was observed for postoperative VAS for pain above 3 points, worse DASH and MEPS scores with a p25 OES. Lower flexion-extension arc of motion, lower levels of satisfaction, lateral ligament insufficency, and the overall presence of complications and implant removal showed moderate correlation with p25 OES.
CONCLUSION
RHR for traumatic complex elbow instability generally provides good long-term outcomes with few complications. Some preoperative characteristics, such as younger age, high-injury trauma, terrible triad, and longitudinal instability, are related to a lower OES with worse overall clinical outcomes. Female sex and younger age may be predictors of an unsatisfactory outcome after an RHR.