Abstract
BACKGROUND: Little is known about the frequency and results of conservative treatment of proximal humerus fractures (PHF) in older individuals. METHODS: Billing data of the BARMER health insurance carrier for all patients of age ≥ 65 for the years 2005-2021 were retrospectively analyzed with multivariable Cox regression models, taking account of the patients' age, sex, and individual comorbidity profiles. The defined primary endpoints were overall survival (OS), major adverse events (MAE), thromboembolic events (TE), and complications of surgery or of trauma. Multivariable p-values for the effect of treatment on all primary endpoints were jointly adjusted using the Bonferroni-Holm method. RESULTS: Of 81 909 patients, 54% were treated conservatively. Conservative treatment was more common in those who received their diagnosis as outpatients (79.5%, vs. 37.2% for inpatients). Operative treatment was associated with significantly longer overall survival (long-term hazard ratio [HR] 0.89, 95% confidence interval [0.86; 0.91]), fewer MAE (0.90 [0.88; 0.92]), and fewer TE (0.89 [0.87; 0.92]), but more complications due to surgery or trauma (1.66 [1,.4; 1.78]; all p < 0.001). By 6 months after diagnosis, 3.1% of the patients who were initially treated conservatively had undergone surgery. Risk factors for failure of conservative treatment were alcohol abuse, obesity, cancer, diabetes mellitus, Parkinson disease, and osteoporosis. CONCLUSION: The conservative treatment of PHF is associated with a lower overall rate of complications due to surgery or trauma, but also with more MAE, more TE, and higher mortality. These findings underline the need for individualized and risk-adjusted treatment recommendations.