Risk factors for delayed healing and multiple revisions in humeral shaft nonunion: a retrospective case series

Abstract Purpose To describe characteristics and outcomes of humeral shaft nonunion. Methods Adult patients diagnosed with humeral shaft nonunion at one Level 1 trauma center from December 2000-July 2020 were retrospectively reviewed. Independent samples t-test was used for continuous variables, and chi squared or fishers exact test for categorical. P < 0.05 was considered significant. Results 59 patients (age 59.83 ± 17.71; 56.14 female; BMI 29.54 ± 8.5) had comorbidities included 32.70% current smoking, 14.30% alcohol use disorder, 21.40% hypothyroidism, 14.30% diabetes. 6 had open fractures; 42 (71.19%) were initially treated nonoperatively. 21 patients were diagnosed with atrophic nonunion, 4 with hypertrophic, and 28 with oligotrophic. 43 nonunions were treated surgically, 11 nonoperatively; the remainder were lost to follow up. 9 had positive cultures at revision. Those with negative revision cultures healed sooner (198 ± 177.54 vs. 451.25 ± 314.377 days, p = 0.034). 41 patients were treated with bone graft at revision. 42 patients (77.78%) required only one revision before radiographic healing; 12 required > 1. Patients with > 1 revision had longer follow-up (846.25 ± 624.75 vs. 358 ± 261.65 days, p < 0.001) and were slower to heal (433 ± 388.75 vs. 202.21 ± 157.38 days, p = 0.028); other factors were not significant on comparison. Conclusion Humeral shaft nonunion remains a complex clinical problem. In this analysis, positive cultures at time of initial revision were found to significantly delay healing. Use of bone graft at time of revision was not significantly associated with faster healing or decreased likelihood of requiring > 1 revision procedure.