Familial Analysis of Nonunion in a Genealogic Population Database Demonstrates No Evidence of Heritable Risk

Abstract

Objective: To determine whether there is evidence of heritable risk for nonunion using a large, state-wide population database. Design: Database. Setting: Level 1 Trauma Center. Population: All Utah residents from 1996 to 2021 who sustained a long bone fracture and their family members were included. Outcomes: The primary outcome was nonunion and the prevalence of nonunion among the patients' first-, second-, and third-degree relatives. The secondary objective was to identify demographic, injury, and socioeconomic risk factors associated with nonunion. Results: In total, 150,263 fractures and 6577 nonunions (4.4%) were identified. This was highly refined to a 1:3 matched cohort of 4667 nonunions of 13,981 fractures for familial clustering analysis. Cox proportional hazards did not demonstrate excessive risk of nonunion among first- (P = 0.863), second- (P = 0.509), and third-degree relatives (P = 0.252). Further analysis of the entire cohort demonstrated that male sex (relative risk [RR] = 1.15; P < 0.001), Medicaid enrollment (RR = 2.64; P < 0.001), open fracture (RR = 2.53; P < 0.001), age group 41–60 years (RR = 1.43; P < 0.001), and a history of obesity (RR = 1.20; P < 0.001) were independent risk factors for nonunion. Conclusions: Our results demonstrate no evidence of heritable risk for nonunion. Independent risk factors for nonunion were male sex, Medicaid enrollment, open fracture, middle age, and a history of obesity. Although it is important to identify modifiable and nonmodifiable risk factors, these results continue to support that the risk of nonunion is multifactorial, relating to injury characteristics, operative techniques, and patient-specific risk factors. Level of Evidence: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.