External validation and revision of the Lafontaine criteria for unstable distal radius fractures: a retrospective study.

Abstract

Background: The Lafontaine criteria are the most commonly cited criteria for predicting unstable distal radius fractures. However, formal validation of the performance of these criteria remains limited. Therefore, we aimed to evaluate the Lafontaine criteria as a diagnostic prediction rule for distal radius fractures, assess the inter-rater reliability and predictive ability of various parameters for distal radius instability, and develop new criteria for fracture instability using reliable and highly predictive factors.

Methods: This retrospective study included 274 adult patients with acute distal radius fractures treated with closed reduction and immobilisation between January 2019 and December 2022. Patients who underwent immediate surgery, were lost to follow-up before 4 weeks, or had unacceptable alignment after reduction were excluded. The Lafontaine criteria were validated using the area under the receiver operating characteristic curve (AUROC). Criteria with an AUROC > 0.7 were considered acceptable. The criteria were updated using risk factors with stronger associations in the multivariable logistic regression analysis, and the inter-rater reliability of potential predictors was evaluated.

Results: The median age of the patients was 63 years; 78% were female. Redisplacement occurred in 39% of the cases. The AUROC for the Lafontaine criteria was 0.65 (95% confidence interval [CI] 0.57-0.74). Multivariable logistic regression showed that age 56-74 years (odds ratio [OR] 3.92, 95% CI 1.82-9.16, p < 0.001) age > 74 years (OR 6.34, 95% CI 2.66-16.2, p < 0.001), associated ulna fracture (OR 1.61, 95% CI 0.92-2.84, p = 0.10), and initial radial shortening > 3 mm (OR 5.78, 95% CI 3.11-11.2, p < 0.001) were the strongest predictive factors of fracture instability. These predictors demonstrated substantial inter-rater reliability, making them suitable for clinical use. Updating the model with these risk factors resulted in an AUROC of 0.74 (95% CI 0.66-0.82).

Conclusions: The performance of the Lafontaine criteria in discriminating unstable distal radius fractures was unacceptable in our study cohort. The updated criteria using age group (< 56 years, 56-74 years, and > 74 years), associated ulnar fractures, and initial radial shortening > 3 mm was found to have moderate discrimination; however, further research is warranted to improve the prediction and measurement reliability of fracture instability.